https://docs.modular.com/search/weekly0.5https://docs.modular.com/weekly0.5https://docs.modular.com/glossary/weekly0.5https://docs.modular.com/glossary/ai/weekly0.5https://docs.modular.com/glossary/ai/attention-mask/weekly0.5https://docs.modular.com/glossary/ai/attention/weekly0.5https://docs.modular.com/glossary/ai/autoregression/weekly0.5https://docs.modular.com/glossary/ai/batching/weekly0.5https://docs.modular.com/glossary/ai/context-encoding/weekly0.5https://docs.modular.com/glossary/ai/continuous-batching/weekly0.5https://docs.modular.com/glossary/ai/disaggregated-inference/weekly0.5https://docs.modular.com/glossary/ai/embedding/weekly0.5https://docs.modular.com/glossary/ai/flash-attention/weekly0.5https://docs.modular.com/glossary/ai/inference-routing/weekly0.5https://docs.modular.com/glossary/ai/kv-cache/weekly0.5https://docs.modular.com/glossary/ai/padding-tokens/weekly0.5https://docs.modular.com/glossary/ai/paged-attention/weekly0.5https://docs.modular.com/glossary/ai/ragged-tensors/weekly0.5https://docs.modular.com/glossary/ai/tokenization/weekly0.5https://docs.modular.com/glossary/ai/transformer/weekly0.5https://docs.modular.com/glossary/gpu/weekly0.5https://docs.modular.com/glossary/gpu/block-index/weekly0.5https://docs.modular.com/glossary/gpu/grid/weekly0.5https://docs.modular.com/glossary/gpu/kernel/weekly0.5https://docs.modular.com/glossary/gpu/memory/weekly0.5https://docs.modular.com/glossary/gpu/occupancy/weekly0.5https://docs.modular.com/glossary/gpu/register/weekly0.5https://docs.modular.com/glossary/gpu/streaming-multiprocessor/weekly0.5https://docs.modular.com/glossary/gpu/thread-block/weekly0.5https://docs.modular.com/glossary/gpu/thread-index/weekly0.5https://docs.modular.com/glossary/gpu/thread/weekly0.5https://docs.modular.com/glossary/gpu/warp/weekly0.5https://docs.modular.com/max/api/c/weekly0.5https://docs.modular.com/max/api/c/common/weekly0.5https://docs.modular.com/max/api/c/context/weekly0.5https://docs.modular.com/max/api/c/model/weekly0.5https://docs.modular.com/max/api/c/tensor/weekly0.5https://docs.modular.com/max/api/c/types/weekly0.5https://docs.modular.com/max/api/kernels/comm/weekly0.5https://docs.modular.com/max/api/kernels/extensibility/weekly0.5https://docs.modular.com/max/api/kernels/kv_cache/weekly0.5https://docs.modular.com/max/api/kernels/linalg/weekly0.5https://docs.modular.com/max/api/kernels/nn/weekly0.5https://docs.modular.com/max/api/kernels/nvml/weekly0.5https://docs.modular.com/max/api/kernels/quantization/weekly0.5https://docs.modular.com/max/api/python/weekly0.5https://docs.modular.com/max/api/python/driver/weekly0.5https://docs.modular.com/max/api/python/dtype/weekly0.5https://docs.modular.com/max/api/python/engine/weekly0.5https://docs.modular.com/max/api/python/entrypoints/weekly0.5https://docs.modular.com/max/api/python/experimental.functional/weekly0.5https://docs.modular.com/max/api/python/experimental.nn.norm/weekly0.5https://docs.modular.com/max/api/python/experimental.nn.rope/weekly0.5https://docs.modular.com/max/api/python/experimental.nn/weekly0.5https://docs.modular.com/max/api/python/experimental.random/weekly0.5https://docs.modular.com/max/api/python/experimental.sharding/weekly0.5https://docs.modular.com/max/api/python/experimental.tensor/weekly0.5https://docs.modular.com/max/api/python/experimental.torch/weekly0.5https://docs.modular.com/max/api/python/experimental/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.accelerator_api/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.accelerator_architecture_name/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.accelerator_count/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.Accelerator/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.Buffer/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.calculate_virtual_device_count_from_cli/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.calculate_virtual_device_count/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.CPU/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.Device/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.DeviceEvent/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.DevicePinnedBuffer/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.devices_exist/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.DeviceSpec/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.DeviceStream/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.DLPackArray/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.enable_all_peer_access/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.get_virtual_device_api/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.get_virtual_device_count/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.get_virtual_device_target_arch/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.is_virtual_device_mode/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.load_devices/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.load_max_buffer/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.NPU/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.scan_available_devices/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.set_virtual_device_api/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.set_virtual_device_count/weekly0.5https://docs.modular.com/max/api/python/generated/max.driver.set_virtual_device_target_arch/weekly0.5https://docs.modular.com/max/api/python/generated/max.dtype.DType/weekly0.5https://docs.modular.com/max/api/python/generated/max.dtype.finfo/weekly0.5https://docs.modular.com/max/api/python/generated/max.engine.CompiledModel/weekly0.5https://docs.modular.com/max/api/python/generated/max.engine.DebugConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.engine.GPUProfilingMode/weekly0.5https://docs.modular.com/max/api/python/generated/max.engine.InferenceSession/weekly0.5https://docs.modular.com/max/api/python/generated/max.engine.LogLevel/weekly0.5https://docs.modular.com/max/api/python/generated/max.engine.Model/weekly0.5https://docs.modular.com/max/api/python/generated/max.engine.TensorSpec/weekly0.5https://docs.modular.com/max/api/python/generated/max.entrypoints.llm.LLM/weekly0.5https://docs.modular.com/max/api/python/generated/max.entrypoints.PipelineConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.CompiledModel/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.Conv2d/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.Embedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.Linear/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.module_dataclass/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.Module/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.ModuleList/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.norm.GemmaRMSNorm/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.norm.GroupNorm/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.norm.LayerNorm/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.norm.rms_norm/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.norm.RMSNorm/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.PinnedDeviceTensor/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.rope.RotaryEmbedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.rope.TransposedRotaryEmbedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.nn.Sequential/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.random.gaussian/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.random.normal/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.random.seed/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.random.set_seed/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.random.uniform/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.ConversionError/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.DeviceMapping/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.DeviceMesh/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.DistributedBufferType/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.DistributedTensorType/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.DistributedType/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.global_shape_from_local/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.local_shard_shape_from_global/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.NamedMapping/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.Partial/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.Placement/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.PlacementMapping/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.ReduceOp/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.Replicated/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.shard_shape/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.sharded_symbolic_dim/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.Sharded/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.SpecEntry/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.sharding.TensorLayout/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.current_realization_context/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.default_device/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.default_dtype/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.defaults_like/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.defaults/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.realization_context/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.RealizationContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.RealizationState/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.tensor.Tensor/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.torch.CustomOpLibrary/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.torch.graph_op/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.torch.max_dtype_to_torch/weekly0.5https://docs.modular.com/max/api/python/generated/max.experimental.torch.torch_dtype_to_max/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.AlgebraicDim/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.BufferType/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.BufferValue/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.BufferValueLike/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.DeviceKind/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.DevicePlacementPolicy/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.DeviceRef/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.Dim/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.DimLike/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.Graph/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.GraphDebugConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.KernelLibrary/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.quantization.BlockParameters/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.quantization.QuantizationConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.quantization.QuantizationEncoding/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.Shape/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.ShapeLike/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.ShardingStrategy/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.StaticDim/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.SymbolicDim/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.TensorType/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.TensorValue/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.TensorValueLike/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.Type/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.Value/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.Weight/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.weights.GGUFWeights/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.weights.load_weights/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.weights.SafetensorWeights/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.weights.WeightData/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.weights.weights_format/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.weights.Weights/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.weights.WeightsAdapter/weekly0.5https://docs.modular.com/max/api/python/generated/max.graph.weights.WeightsFormat/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Allreduce/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.AttentionMaskVariant/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.AttentionWithRope/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.DistributedAttentionImpl/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.GGUFQAttentionWithRope/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.GPTQAttentionWithRope/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.LatentAttentionWithRope/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.MHAMaskVariant/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.MultiheadAttention/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.num_heads_for_device/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.RaggedAttention/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.TensorParallelAttentionWithRope/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.attention.TensorParallelLatentAttentionWithRope/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.AttentionWithRopeAndLoRA/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.build_max_lengths_tensor/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.clamp/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.ColumnParallelLinear/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.ConstantLayerNorm/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Conv1D/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Conv2d/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Conv3D/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.ConvTranspose1d/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.DistributedTransformer/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.DistributedTransformerBlock/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.DynamicRotaryEmbedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Embedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.forward_moe_sharded_layers/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.GPTQLinear/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.GroupNorm/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.hooks.PrintHook/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Identity/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.InputScaleSpec/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.AttentionDispatchResolver/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.build_max_lengths_tensor/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.compute_max_seq_len_fitting_in_cache/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.compute_num_device_blocks/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.compute_num_host_blocks/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.estimated_memory_size/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.KVCacheBuffer/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.KVCacheInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.KVCacheInputsPerDevice/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.KVCacheMetrics/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.KVCacheParamInterface/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.KVCacheParams/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.KVCacheQuantizationConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.KVConnectorType/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.MultiKVCacheParams/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.kv_cache.PagedCacheValues/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Layer/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.LayerList/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.LayerNorm/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Linear/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.LinearLoRA/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.LinearScalingParams/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Llama3RopeScalingParams/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Llama3RotaryEmbedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.LongRoPERotaryEmbedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.LongRoPEScalingParams/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.MinPSampler/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.MLP/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Module/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.MoE/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.MoEGate/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.MoEQuantized/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.QuantConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.QuantFormat/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.RejectionSampler/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.RejectionSamplerWithResiduals/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.ReturnHiddenStates/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.ReturnLogits/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.RMSNorm/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.RotaryEmbedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.ScaleGranularity/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.ScaleOrigin/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Sequential/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Shardable/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Signals/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.split_batch_replicated/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.split_batch/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.StackedLinear/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.SupportsLoRA/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.Transformer/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.TransformerBlock/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.VocabParallelEmbedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.WeightNormConvTranspose1d/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.WeightScaleSpec/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.YarnRotaryEmbedding/weekly0.5https://docs.modular.com/max/api/python/generated/max.nn.YarnScalingParams/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.ADAPTER_CONFIG_FILE/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.AudioGenerationConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.core.SpecDecodingState/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.core.validate_aspect_ratio_args/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.core.validate_image_grid_thw_args/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.core.validate_image_shape_5d/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.core.validate_initial_prompt_has_image/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.core.validate_only_one_image/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.core.validate_requires_vision_context/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.core.validate_vision_position_ids/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.download_weight_files/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.EmbeddingsPipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.GenerateMixin/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.IdentityPipelineTokenizer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.is_float4_encoding/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.available_port/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.DummyKVCache/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.IncrementCacheLengthsProcessor/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.InsufficientBlocksError/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.KVTransferEngine/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.KVTransferEngineMetadata/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.load_kv_manager/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.load_multi_kv_managers/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.PagedKVCacheManager/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.kv_cache.TransferReqData/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.KVCacheConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.AlwaysSignalBuffersMixin/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.build_eos_tracker_for_request/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.CompilationTimer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.convert_max_config_value/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.deep_merge_max_configs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.DenoisingCacheConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.EmbeddingsPipelineType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.float32_array_to_buffer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.float32_to_bfloat16_as_uint16/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.generate_local_model_path/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.get_default_max_config_file_section_name/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.HuggingFaceRepo/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.interfaces.ArchConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.interfaces.ArchConfigWithAttentionKVCache/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.interfaces.ArchConfigWithKVCache/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.interfaces.DiffusionPipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.KVConnectorConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.log_probabilities.compute_log_probabilities_ragged/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.log_probabilities.log_probabilities_ragged_graph/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.LoRAManager/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.LoRARequestProcessor/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.max_tokens_to_generate/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.MAXConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.MAXModelConfigBase/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.ModelManifest/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.OverlapTextGenerationPipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.parse_quant_config/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.PipelineModelWithKVCache/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.PipelineRuntimeConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.PixelGenerationTokenizer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.registry.PIPELINE_REGISTRY/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.registry.PipelineModelType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.registry.PipelineRegistry/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.registry.PipelineTypes/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.registry.SupportedArchitecture/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.rejection_sampler_with_residuals/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.rejection_sampler/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.resolve_max_config_inheritance/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.StandaloneSpeculativeDecodingPipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.token_sampler/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.try_to_load_from_cache/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.UnifiedEagleOutputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.validate_hf_repo_access/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.lib.WeightPathParser/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.LoRAConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.MAXModelConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.MemoryEstimator/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.ComponentModel/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.DenoisingCacheConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.DenoisingCacheState/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.fbcache_conditional_execution/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.FirstBlockCache/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.FirstBlockCacheState/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.run_denoising_step/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.TaylorSeer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.TaylorSeerState/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.base.TensorStruct/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.dataprocessing.batch_padded_tokens_and_mask/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.dataprocessing.causal_attention_mask_with_token_mask/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.dataprocessing.causal_attention_mask/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.dataprocessing.collate_batch/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.dataprocessing.max_tokens_to_generate/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.dataprocessing.PaddingDirection/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.kv_cache_config.KVCacheConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.kv_cache_config.KVConnectorConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.AudioGenerationContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.AudioGenerationInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.AudioGenerationMetadata/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.AudioGenerationOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.AudioGenerationRequest/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.BaseContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.BaseContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.BatchLogitsProcessor/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.BatchProcessorInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.BatchType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.DUMMY_REQUEST_ID/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.EmbeddingsContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.EmbeddingsGenerationContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.EmbeddingsGenerationInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.EmbeddingsGenerationOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.EOSTracker/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.GenerationOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.GenerationStatus/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ImageContentPart/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ImageMetadata/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.InputModality/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LogitsProcessor/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LogProbabilities/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LORA_REQUEST_ENDPOINT/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LORA_RESPONSE_ENDPOINT/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LoRAOperation/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LoRARequest/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LoRAResponse/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LoRAStatus/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.LoRAType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.MessageContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.msgpack_numpy_decoder/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.msgpack_numpy_encoder/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.OpenResponsesRequest/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ParsedReasoningDelta/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ParsedToolCall/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ParsedToolCallDelta/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ParsedToolResponse/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.AudioGenerationContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.AudioGenerationInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.AudioGenerationMetadata/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.AudioGenerationOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.AudioGenerationRequest/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.BatchType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.EmbeddingsContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.EmbeddingsGenerationContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.EmbeddingsGenerationInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.EmbeddingsGenerationOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.ImageContentPart/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.ImageMetadata/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.MessageContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.PixelGenerationContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.PixelGenerationContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.PixelGenerationInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.SpecDecodingState/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextContentPart/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationRequest/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationRequestFunction/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationRequestMessage/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationRequestTool/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.TextGenerationResponseFormat/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.VideoContentPart/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.VLMContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.pipeline_variants.VLMTextGenerationContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.Pipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PipelineInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PipelineInputsType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PipelineOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PipelineOutputsDict/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PipelineOutputType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PipelinesFactory/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PipelineTask/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PipelineTokenizer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PixelGenerationContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PixelGenerationContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.PixelGenerationInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ProcessorInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.Range/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ReasoningParser/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ReasoningSpan/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.Request/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.RequestID/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.RequestType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.SamplingParams/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.SamplingParamsGenerationConfigDefaults/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.SamplingParamsInput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.SharedMemoryArray/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.SpecDecodingState/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextContentPart/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationRequest/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationRequestFunction/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationRequestMessage/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationRequestTool/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TextGenerationResponseFormat/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TokenBuffer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TokenizerEncoded/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.TokenSlice/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.ToolParser/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.UnboundContextType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.VideoContentPart/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.types.VLMTextGenerationContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.AUTO_CAST_ENV_VAR/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.auto_cast_weights_from_env/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.download_weight_files/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.generate_local_model_path/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.HuggingFaceRepo/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.is_diffusion_pipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.try_to_load_from_cache/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.validate_hf_repo_access/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.modeling.weights.WeightPathParser/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.ModelInputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.ModelOutputs/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.parse_supported_encoding_from_file_name/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.PipelineConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.PipelineModel/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.PipelineRole/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.PixelContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.PixelGenerationPipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.PrependPromptSpeechTokens/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.ProfilingConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.PrometheusMetricsMode/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.RepoType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.AllowedToolChoice/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.AssistantMessage/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.DeveloperMessage/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.DUMMY_REQUEST_ID/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.Error/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.FastAPIRequestProtocol/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.FunctionCall/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.FunctionCallOutput/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.FunctionCallStatus/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.FunctionTool/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.FunctionToolChoice/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.FunctionToolParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ImageDetail/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.IncludeEnum/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.IncompleteDetails/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.InputContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.InputFileContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.InputImageContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.InputMessage/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.InputTextContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.InputTokensDetails/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.InputVideoContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ItemReferenceParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.JsonObjectField/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.JsonObjectParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.JsonSchemaField/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.JsonSchemaParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.LogProb/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.Message/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.MessageRole/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.MessageStatus/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.OpenResponsesRequest/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.OpenResponsesRequestBody/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.OutputContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.OutputImageContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.OutputTextContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.OutputTokensDetails/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.OutputVideoContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.provider_options.GeneratedMediaResponseFormat/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.provider_options.ImageProviderOptions/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.provider_options.MaxProviderOptions/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.provider_options.PixelProviderOptionsBase/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.provider_options.ProviderOptions/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.provider_options.VideoProviderOptions/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ReasoningBody/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ReasoningEffortEnum/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ReasoningParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ReasoningReference/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ReasoningReferenceParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ReasoningSummaryContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ReasoningSummaryEnum/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.RefusalContent/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.Request/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.RequestID/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.RequestType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ResponseFormat/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ResponseFormatParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ResponseResource/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ServiceTierEnum/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.StreamOptionsParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.SystemMessage/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.TextField/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.TextParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ToolChoice/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.ToolChoiceValueEnum/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.TopLogProb/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.TruncationEnum/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.UrlCitationBody/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.UrlCitationParam/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.Usage/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.UserMessage/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.request.VerbosityEnum/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.RopeType/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.SamplingConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.SpeculativeConfig/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.SpeechTokenGenerationPipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.supported_encoding_dtype/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.supported_encoding_quantization/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.supported_encoding_supported_devices/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.supported_encoding_supported_on/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.SupportedEncoding/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.TextAndVisionContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.TextAndVisionTokenizer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.TextContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.TextGenerationPipeline/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.TextGenerationPipelineInterface/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.TextTokenizer/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.TTSContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.pipelines.upper_bounded_default/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.cpu.collect_pids_for_port/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.cpu.CPUMetrics/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.cpu.CPUMetricsCollector/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.gpu.BackgroundRecorder/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.gpu.ClockStats/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.gpu.GPUDiagContext/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.gpu.GPUStats/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.gpu.HARDWARE_THROTTLE_REASONS/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.gpu.MemoryStats/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.gpu.ThrottleReason/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.gpu.UtilizationStats/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.is_profiling_enabled/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.set_gpu_profiling_state/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.traced/weekly0.5https://docs.modular.com/max/api/python/generated/max.profiler.Tracer/weekly0.5https://docs.modular.com/max/api/python/generated/max.serve.queue.drain_queue/weekly0.5https://docs.modular.com/max/api/python/generated/max.serve.queue.get_blocking/weekly0.5https://docs.modular.com/max/api/python/generated/max.serve.queue.MAXPullQueue/weekly0.5https://docs.modular.com/max/api/python/generated/max.serve.queue.MAXPushQueue/weekly0.5https://docs.modular.com/max/api/python/generated/max.serve.scheduler_result.SchedulerResult/weekly0.5https://docs.modular.com/max/api/python/generated/max.serve.scheduler.interface.Scheduler/weekly0.5https://docs.modular.com/max/api/python/graph.ops/weekly0.5https://docs.modular.com/max/api/python/graph.quantization/weekly0.5https://docs.modular.com/max/api/python/graph.weights/weekly0.5https://docs.modular.com/max/api/python/graph/weekly0.5https://docs.modular.com/max/api/python/nn.attention/weekly0.5https://docs.modular.com/max/api/python/nn.kernels/weekly0.5https://docs.modular.com/max/api/python/nn.kv_cache/weekly0.5https://docs.modular.com/max/api/python/nn/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.bert/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.deepseekV2/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.deepseekV3_2/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.deepseekV3_nextn/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.deepseekV3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.dflash_llama3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.eagle_llama3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.eagle3_deepseekV3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.exaone_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.exaone/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.flux2/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.gemma3_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.gemma3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.gemma3multimodal_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.gemma3multimodal/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.gemma4/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.gpt_oss_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.gpt_oss/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.granite_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.granite/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.idefics3_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.idefics3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.internvl/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.kimik2_5/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.lfm2/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.llama3_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.llama3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.mamba/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.minimax_m2/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.mistral/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.mistral3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.mpnet_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.mpnet/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.olmo_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.olmo/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.olmo2_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.olmo2/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.olmo3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.phi3_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.phi3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.pixtral_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.pixtral/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen_image_edit/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen_image/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen2_5vl/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen2/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen3_5/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen3_embedding_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen3_embedding/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.qwen3vl_moe/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.step3p5/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.unified_dflash_llama3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.unified_eagle_llama3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.unified_mtp_deepseekV3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.wan/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures.z_image_modulev3/weekly0.5https://docs.modular.com/max/api/python/pipelines.architectures/weekly0.5https://docs.modular.com/max/api/python/pipelines.core/weekly0.5https://docs.modular.com/max/api/python/pipelines.kv_cache/weekly0.5https://docs.modular.com/max/api/python/pipelines.lib.interfaces/weekly0.5https://docs.modular.com/max/api/python/pipelines.lib.log_probabilities/weekly0.5https://docs.modular.com/max/api/python/pipelines.lib.registry/weekly0.5https://docs.modular.com/max/api/python/pipelines.lib/weekly0.5https://docs.modular.com/max/api/python/pipelines.modeling.base/weekly0.5https://docs.modular.com/max/api/python/pipelines.modeling.dataprocessing/weekly0.5https://docs.modular.com/max/api/python/pipelines.modeling.kv_cache_config/weekly0.5https://docs.modular.com/max/api/python/pipelines.modeling.types.pipeline_variants/weekly0.5https://docs.modular.com/max/api/python/pipelines.modeling.types/weekly0.5https://docs.modular.com/max/api/python/pipelines.modeling.weights/weekly0.5https://docs.modular.com/max/api/python/pipelines.request.provider_options/weekly0.5https://docs.modular.com/max/api/python/pipelines.request/weekly0.5https://docs.modular.com/max/api/python/pipelines/weekly0.5https://docs.modular.com/max/api/python/profiler.cpu/weekly0.5https://docs.modular.com/max/api/python/profiler.gpu/weekly0.5https://docs.modular.com/max/api/python/profiler/weekly0.5https://docs.modular.com/max/api/python/serve.queue/weekly0.5https://docs.modular.com/max/api/python/serve.scheduler_result/weekly0.5https://docs.modular.com/max/api/python/serve.scheduler.interface/weekly0.5https://docs.modular.com/max/changelog/weekly0.5https://docs.modular.com/max/changelog/nightly/weekly0.5https://docs.modular.com/max/changelog/v0.24.1.1/weekly0.5https://docs.modular.com/max/changelog/v0.24.1/weekly0.5https://docs.modular.com/max/changelog/v0.24.2.1/weekly0.5https://docs.modular.com/max/changelog/v0.24.2/weekly0.5https://docs.modular.com/max/changelog/v0.24.3/weekly0.5https://docs.modular.com/max/changelog/v0.24.4/weekly0.5https://docs.modular.com/max/changelog/v0.24.5/weekly0.5https://docs.modular.com/max/changelog/v0.24.6/weekly0.5https://docs.modular.com/max/changelog/v0.25.1.1/weekly0.5https://docs.modular.com/max/changelog/v0.25.1/weekly0.5https://docs.modular.com/max/changelog/v0.25.2/weekly0.5https://docs.modular.com/max/changelog/v0.25.3/weekly0.5https://docs.modular.com/max/changelog/v0.25.4/weekly0.5https://docs.modular.com/max/changelog/v0.25.5/weekly0.5https://docs.modular.com/max/changelog/v0.25.6.1/weekly0.5https://docs.modular.com/max/changelog/v0.25.6/weekly0.5https://docs.modular.com/max/changelog/v0.25.7/weekly0.5https://docs.modular.com/max/changelog/v0.26.1/weekly0.5https://docs.modular.com/max/changelog/v0.26.2/weekly0.5https://docs.modular.com/max/changelog/v0.26.3/weekly0.5https://docs.modular.com/max/cli/weekly0.5https://docs.modular.com/max/cli/benchmark/weekly0.5https://docs.modular.com/max/cli/encode/weekly0.5https://docs.modular.com/max/cli/generate/weekly0.5https://docs.modular.com/max/cli/list/weekly0.5https://docs.modular.com/max/cli/serve/weekly0.5https://docs.modular.com/max/cli/warm-cache/weekly0.5https://docs.modular.com/max/coding-assistants/weekly0.5https://docs.modular.com/max/container/weekly0.5https://docs.modular.com/max/deploy/benchmark/weekly0.5https://docs.modular.com/max/deploy/cloud/weekly0.5https://docs.modular.com/max/deploy/local-to-cloud/weekly0.5https://docs.modular.com/max/develop/weekly0.5https://docs.modular.com/max/develop/basic-ops/weekly0.5https://docs.modular.com/max/develop/broadcasting/weekly0.5https://docs.modular.com/max/develop/build-custom-ops/weekly0.5https://docs.modular.com/max/develop/custom-kernels-pytorch/weekly0.5https://docs.modular.com/max/develop/custom-ops-matmul/weekly0.5https://docs.modular.com/max/develop/custom-ops/weekly0.5https://docs.modular.com/max/develop/debugging-accuracy/weekly0.5https://docs.modular.com/max/develop/debugging-errors/weekly0.5https://docs.modular.com/max/develop/debugging-gpu/weekly0.5https://docs.modular.com/max/develop/debugging-tracing/weekly0.5https://docs.modular.com/max/develop/debugging/weekly0.5https://docs.modular.com/max/develop/dtypes/weekly0.5https://docs.modular.com/max/develop/eager-execution/weekly0.5https://docs.modular.com/max/develop/graph/weekly0.5https://docs.modular.com/max/develop/indexing/weekly0.5https://docs.modular.com/max/develop/max-pipeline-bring-your-own-model/weekly0.5https://docs.modular.com/max/develop/model-bringup-workflow/weekly0.5https://docs.modular.com/max/develop/modules/weekly0.5https://docs.modular.com/max/develop/pipelines/weekly0.5https://docs.modular.com/max/develop/serve-custom-model-architectures/weekly0.5https://docs.modular.com/max/develop/tensor-realization/weekly0.5https://docs.modular.com/max/develop/tensors/weekly0.5https://docs.modular.com/max/environment-variables/weekly0.5https://docs.modular.com/max/faq/weekly0.5https://docs.modular.com/max/get-started/weekly0.5https://docs.modular.com/max/gpu-system-profiling/weekly0.5https://docs.modular.com/max/graph/quantize/weekly0.5https://docs.modular.com/max/inference/embeddings/weekly0.5https://docs.modular.com/max/inference/image-generation/weekly0.5https://docs.modular.com/max/inference/image-to-text/weekly0.5https://docs.modular.com/max/inference/text-to-text/weekly0.5https://docs.modular.com/max/inference/video-generation/weekly0.5https://docs.modular.com/max/intro/weekly0.5https://docs.modular.com/max/models/weekly0.5https://docs.modular.com/max/nightly-changelog/weekly0.5https://docs.modular.com/max/packages/weekly0.5https://docs.modular.com/max/rest-api/weekly0.5https://docs.modular.com/max/serve/function-calling/weekly0.5https://docs.modular.com/max/serve/lora-adapters/weekly0.5https://docs.modular.com/max/serve/offline-inference/weekly0.5https://docs.modular.com/max/serve/prefix-caching/weekly0.5https://docs.modular.com/max/serve/speculative-decoding/weekly0.5https://docs.modular.com/max/serve/structured-output/weekly0.5https://docs.modular.com/pixi/weekly0.5