List of all items
Structs
- comm::SharedMemoryComm
- comm::SharedMemoryGroup
- comm::TcpComm
- ops::flash_attention::AttentionWorkspace
- ops::flash_attention::DeterministicConfig
- ops::flash_attention::FlashAttentionConfig
- ops::flash_attention::HierarchicalFlashAttention
- ops::flash_attention::MaskCacheConfig
- ops::flash_attention::StrictOrderIterator
- ops::flash_attention_v3::FlashAttention3
- ops::flash_attention_v3::FlashAttention3Config
- ops::kv_compression::CompressedKV
- ops::kv_compression::KVCacheCompressor
- ops::mamba::HybridLayer
- ops::mamba::MambaBlock
- ops::mamba::MambaConfig
- ops::mamba::MambaParameters
- ops::mamba::MambaState
- ops::mla::CompressedKVCache
- ops::mla::MultiHeadLatentAttention
- ops::paged_attention::BlockManager
- ops::paged_attention::BlockTable
- ops::paged_attention::KVBlock
- ops::paged_attention::KVBlockIterator
- ops::paged_attention::KVBlockRef
- ops::paged_attention::PagedAttention
- ops::paged_attention::PagedKVCache
- ops::ring_attention::RingAttention
- ops::ring_attention::RingAttentionConfig
- ops::softmax::LogSpaceSoftmax
- ops::softmax::TensorLogOps
- ops::sparse_attention::SparseAttention
- ops::sparse_attention::SparseAttentionConfig
- ops::sparse_attention::SparseSelection
- ops::speculative_decoding::PredictionConfig
- ops::speculative_decoding::SpeculativeCandidates
- ops::speculative_decoding::SpeculativeDecoder
- ops::speculative_decoding::SpeculativeToken
- ops::speculative_decoding::SpeculativeTree
- ops::speculative_decoding::SpeculativeVerification
- ops::speculative_decoding::TreeConfig
- ops::stable_accumulator::AccumulatorConfig
- ops::stable_accumulator::HierarchicalAccumulator
- ops::stable_accumulator::KahanAccumulator
- ops::stable_accumulator::StableAccumulator
- ops::stable_accumulator::StableRowState
- types::AttentionConfig
- types::PagedAttentionConfig
Enums
- comm::CommError
- ops::kv_compression::CompressionMethod
- ops::kv_compression::KVLayout
- ops::mamba::HybridStrategy
- ops::ring_attention::CommBackend
- ops::sparse_attention::SparsityPattern
- ops::speculative_decoding::PredictionHeadType
- ops::speculative_decoding::VerificationStrategy
- types::KernelPrecision
Traits
Functions
- backend::select_device
- device::default_device
- device::device_for
- ops::softmax::log_add_exp
- ops::softmax::log_sum_exp
- ops::softmax::log_sum_exp_kahan
Type Aliases
- backend::DefaultBackend
- comm::CommResult
- device::DefaultDevice
- ops::flash_attention::HierarchicalFlashConfig
- ops::stable_accumulator::KahanSum
- ops::stable_accumulator::OutputAccumulator
- types::ConfigResult