LICENSE
README.md
pyproject.toml
setup.py
python/freetoken/__init__.py
python/freetoken/__main__.py
python/freetoken/cache_report.py
python/freetoken/cli.py
python/freetoken/control_cli.py
python/freetoken/core.py
python/freetoken/env.py
python/freetoken/launch.py
python/freetoken/version.py
python/freetoken.egg-info/PKG-INFO
python/freetoken.egg-info/SOURCES.txt
python/freetoken.egg-info/dependency_links.txt
python/freetoken.egg-info/entry_points.txt
python/freetoken.egg-info/requires.txt
python/freetoken.egg-info/top_level.txt
python/freetoken/attention/__init__.py
python/freetoken/attention/base.py
python/freetoken/attention/dsa.py
python/freetoken/attention/dsa_indexer.py
python/freetoken/attention/dsv4_compress.py
python/freetoken/attention/dsv4_indexer.py
python/freetoken/attention/dsv4_sparse.py
python/freetoken/attention/fa.py
python/freetoken/attention/fi.py
python/freetoken/attention/linear.py
python/freetoken/attention/m3_sparse.py
python/freetoken/attention/triton.py
python/freetoken/attention/trtllm.py
python/freetoken/attention/utils.py
python/freetoken/benchmark/client.py
python/freetoken/benchmark/perf.py
python/freetoken/checkpoint/__init__.py
python/freetoken/checkpoint/__main__.py
python/freetoken/checkpoint/convert.py
python/freetoken/checkpoint/ftw.py
python/freetoken/daemon/__init__.py
python/freetoken/daemon/__main__.py
python/freetoken/daemon/accounting.py
python/freetoken/daemon/app.py
python/freetoken/daemon/checkpoint.py
python/freetoken/daemon/client.py
python/freetoken/daemon/logfmt.py
python/freetoken/daemon/logring.py
python/freetoken/daemon/metrics.py
python/freetoken/daemon/osproc.py
python/freetoken/daemon/pidfile.py
python/freetoken/daemon/proxy.py
python/freetoken/daemon/serve_manager.py
python/freetoken/daemon/server.py
python/freetoken/daemon/tailer.py
python/freetoken/daemon/version.py
python/freetoken/distributed/__init__.py
python/freetoken/distributed/impl.py
python/freetoken/distributed/info.py
python/freetoken/engine/__init__.py
python/freetoken/engine/cache_budget.py
python/freetoken/engine/config.py
python/freetoken/engine/engine.py
python/freetoken/engine/graph.py
python/freetoken/engine/sample.py
python/freetoken/kernel/__init__.py
python/freetoken/kernel/__main__.py
python/freetoken/kernel/_toolchain.py
python/freetoken/kernel/aot.py
python/freetoken/kernel/aot_models.py
python/freetoken/kernel/backend.py
python/freetoken/kernel/batch_memcpy.py
python/freetoken/kernel/causal_conv1d.py
python/freetoken/kernel/fast_index_copy.py
python/freetoken/kernel/gguf.py
python/freetoken/kernel/index.py
python/freetoken/kernel/moe_impl.py
python/freetoken/kernel/pinned.py
python/freetoken/kernel/pynccl.py
python/freetoken/kernel/radix.py
python/freetoken/kernel/store.py
python/freetoken/kernel/tensor.py
python/freetoken/kernel/utils.py
python/freetoken/kernel/csrc/pinned_tensor.cpp
python/freetoken/kernel/csrc/cpu_moe/cpu_moe_ext.cpp
python/freetoken/kernel/csrc/gguf/dequantize.cuh
python/freetoken/kernel/csrc/gguf/dispatch.h
python/freetoken/kernel/csrc/gguf/ggml-common.h
python/freetoken/kernel/csrc/gguf/gguf_kernel.cu
python/freetoken/kernel/csrc/gguf/mmq.cuh
python/freetoken/kernel/csrc/gguf/mmvq.cuh
python/freetoken/kernel/csrc/gguf/moe.cuh
python/freetoken/kernel/csrc/gguf/moe_vec.cuh
python/freetoken/kernel/csrc/gguf/vecdotq.cuh
python/freetoken/kernel/csrc/include/freetoken/nccl227.h
python/freetoken/kernel/csrc/include/freetoken/tensor.h
python/freetoken/kernel/csrc/include/freetoken/utils.cuh
python/freetoken/kernel/csrc/include/freetoken/utils.h
python/freetoken/kernel/csrc/include/freetoken/warp.cuh
python/freetoken/kernel/csrc/jit/batch_memcpy.cuh
python/freetoken/kernel/csrc/jit/fast_index_copy.cuh
python/freetoken/kernel/csrc/jit/index.cu
python/freetoken/kernel/csrc/jit/store.cu
python/freetoken/kernel/csrc/src/pynccl.cu
python/freetoken/kernel/csrc/src/radix.cpp
python/freetoken/kernel/csrc/src/tensor.cpp
python/freetoken/kernel/fla/__init__.py
python/freetoken/kernel/fla/chunk.py
python/freetoken/kernel/fla/chunk_delta_h.py
python/freetoken/kernel/fla/chunk_fwd.py
python/freetoken/kernel/fla/chunk_o.py
python/freetoken/kernel/fla/cumsum.py
python/freetoken/kernel/fla/fused_sigmoid_gating_recurrent.py
python/freetoken/kernel/fla/index.py
python/freetoken/kernel/fla/l2norm.py
python/freetoken/kernel/fla/layernorm_gated.py
python/freetoken/kernel/fla/op.py
python/freetoken/kernel/fla/utils.py
python/freetoken/kernel/fla/wy_fast.py
python/freetoken/kernel/triton/activation.py
python/freetoken/kernel/triton/attention.py
python/freetoken/kernel/triton/autotune_cache.py
python/freetoken/kernel/triton/causal_conv1d_triton.py
python/freetoken/kernel/triton/decode_moe.py
python/freetoken/kernel/triton/df11.py
python/freetoken/kernel/triton/df11_decode.py
python/freetoken/kernel/triton/e4m3_compat.py
python/freetoken/kernel/triton/fp8_block_linear.py
python/freetoken/kernel/triton/fp8_blockscale_moe.py
python/freetoken/kernel/triton/fp8_pertensor_linear.py
python/freetoken/kernel/triton/fused_moe.py
python/freetoken/kernel/triton/gemma4_fused.py
python/freetoken/kernel/triton/glm_dsa_sparse.py
python/freetoken/kernel/triton/minimax_m3_sparse.py
python/freetoken/kernel/triton/moe_align.py
python/freetoken/kernel/triton/mxfp4_moe.py
python/freetoken/kernel/triton/mxfp8_linear.py
python/freetoken/kernel/triton/norm.py
python/freetoken/kernel/triton/nvfp4_dequant.py
python/freetoken/kernel/triton/nvfp4_fused_moe.py
python/freetoken/kernel/triton/nvfp4_linear.py
python/freetoken/kernel/triton/rope.py
python/freetoken/kernel/triton/sampling.py
python/freetoken/kernel/triton/dsv4/bf16_linear.py
python/freetoken/kernel/triton/dsv4/compress.py
python/freetoken/kernel/triton/dsv4/fp8_linear.py
python/freetoken/kernel/triton/dsv4/fused_moe.py
python/freetoken/kernel/triton/dsv4/hadamard.py
python/freetoken/kernel/triton/dsv4/hc.py
python/freetoken/kernel/triton/dsv4/indexer.py
python/freetoken/kernel/triton/dsv4/norm.py
python/freetoken/kernel/triton/dsv4/rope.py
python/freetoken/kernel/triton/dsv4/sinkhorn.py
python/freetoken/kernel/triton/dsv4/sparse_attn.py
python/freetoken/kernel/triton/dsv4/swiglu.py
python/freetoken/kvcache/__init__.py
python/freetoken/kvcache/base.py
python/freetoken/kvcache/bsa_pool.py
python/freetoken/kvcache/cache_status.py
python/freetoken/kvcache/dsa_pool.py
python/freetoken/kvcache/dsv4_cost_model.py
python/freetoken/kvcache/dsv4_paged_pool.py
python/freetoken/kvcache/hybrid_radix_cache.py
python/freetoken/kvcache/hybrid_swa_pool.py
python/freetoken/kvcache/linear_state_pool.py
python/freetoken/kvcache/mha_pool.py
python/freetoken/kvcache/naive_cache.py
python/freetoken/kvcache/radix_cache.py
python/freetoken/kvcache/swa_radix_cache.py
python/freetoken/layers/__init__.py
python/freetoken/layers/activation.py
python/freetoken/layers/base.py
python/freetoken/layers/embedding.py
python/freetoken/layers/gguf.py
python/freetoken/layers/linear.py
python/freetoken/layers/moe.py
python/freetoken/layers/norm.py
python/freetoken/layers/rotary.py
python/freetoken/llm/__init__.py
python/freetoken/llm/llm.py
python/freetoken/message/__init__.py
python/freetoken/message/backend.py
python/freetoken/message/frontend.py
python/freetoken/message/tokenizer.py
python/freetoken/message/utils.py
python/freetoken/models/__init__.py
python/freetoken/models/blocks.py
python/freetoken/models/config.py
python/freetoken/models/loader.py
python/freetoken/models/nvfp4_banks.py
python/freetoken/models/register.py
python/freetoken/models/weight.py
python/freetoken/models/deepseek_v4/__init__.py
python/freetoken/models/deepseek_v4/args.py
python/freetoken/models/deepseek_v4/attention.py
python/freetoken/models/deepseek_v4/compress.py
python/freetoken/models/deepseek_v4/config.py
python/freetoken/models/deepseek_v4/layers.py
python/freetoken/models/deepseek_v4/model.py
python/freetoken/models/deepseek_v4/moe.py
python/freetoken/models/deepseek_v4/ops.py
python/freetoken/models/deepseek_v4/weight.py
python/freetoken/models/gemma4/__init__.py
python/freetoken/models/gemma4/attention.py
python/freetoken/models/gemma4/config.py
python/freetoken/models/gemma4/gguf.py
python/freetoken/models/gemma4/model.py
python/freetoken/models/gemma4/moe.py
python/freetoken/models/gemma4/vision.py
python/freetoken/models/gemma4/weight.py
python/freetoken/models/gguf/__init__.py
python/freetoken/models/gguf/config.py
python/freetoken/models/gguf/dequant.py
python/freetoken/models/gguf/reader.py
python/freetoken/models/gguf/tokenizer.py
python/freetoken/models/glm4_moe/__init__.py
python/freetoken/models/glm4_moe/attention.py
python/freetoken/models/glm4_moe/config.py
python/freetoken/models/glm4_moe/df11_embedding.py
python/freetoken/models/glm4_moe/df11_linear.py
python/freetoken/models/glm4_moe/mlp.py
python/freetoken/models/glm4_moe/model.py
python/freetoken/models/glm4_moe/moe.py
python/freetoken/models/glm4_moe/nvfp4_linear.py
python/freetoken/models/glm4_moe/weight.py
python/freetoken/models/glm_moe_dsa/__init__.py
python/freetoken/models/glm_moe_dsa/args.py
python/freetoken/models/glm_moe_dsa/attention.py
python/freetoken/models/glm_moe_dsa/config.py
python/freetoken/models/glm_moe_dsa/mlp.py
python/freetoken/models/glm_moe_dsa/model.py
python/freetoken/models/glm_moe_dsa/moe.py
python/freetoken/models/glm_moe_dsa/weight.py
python/freetoken/models/gpt_oss/__init__.py
python/freetoken/models/gpt_oss/attention.py
python/freetoken/models/gpt_oss/config.py
python/freetoken/models/gpt_oss/model.py
python/freetoken/models/gpt_oss/moe.py
python/freetoken/models/gpt_oss/weight.py
python/freetoken/models/llama/__init__.py
python/freetoken/models/llama/attention.py
python/freetoken/models/llama/config.py
python/freetoken/models/llama/model.py
python/freetoken/models/llama/weight.py
python/freetoken/models/minimax_m2/__init__.py
python/freetoken/models/minimax_m2/attention.py
python/freetoken/models/minimax_m2/config.py
python/freetoken/models/minimax_m2/model.py
python/freetoken/models/minimax_m2/moe.py
python/freetoken/models/minimax_m2/weight.py
python/freetoken/models/minimax_m3/__init__.py
python/freetoken/models/minimax_m3/args.py
python/freetoken/models/minimax_m3/attention.py
python/freetoken/models/minimax_m3/config.py
python/freetoken/models/minimax_m3/mlp.py
python/freetoken/models/minimax_m3/model.py
python/freetoken/models/minimax_m3/moe.py
python/freetoken/models/minimax_m3/weight.py
python/freetoken/models/mistral/__init__.py
python/freetoken/models/mistral/attention.py
python/freetoken/models/mistral/config.py
python/freetoken/models/mistral/model.py
python/freetoken/models/mistral/weight.py
python/freetoken/models/qwen2/__init__.py
python/freetoken/models/qwen2/attention.py
python/freetoken/models/qwen2/config.py
python/freetoken/models/qwen2/model.py
python/freetoken/models/qwen2/weight.py
python/freetoken/models/qwen3/__init__.py
python/freetoken/models/qwen3/attention.py
python/freetoken/models/qwen3/config.py
python/freetoken/models/qwen3/model.py
python/freetoken/models/qwen3/weight.py
python/freetoken/models/qwen3_5_moe/__init__.py
python/freetoken/models/qwen3_5_moe/attention.py
python/freetoken/models/qwen3_5_moe/config.py
python/freetoken/models/qwen3_5_moe/gdn.py
python/freetoken/models/qwen3_5_moe/gdn_kernels.py
python/freetoken/models/qwen3_5_moe/gdn_reference.py
python/freetoken/models/qwen3_5_moe/model.py
python/freetoken/models/qwen3_5_moe/moe.py
python/freetoken/models/qwen3_5_moe/quant_linear.py
python/freetoken/models/qwen3_5_moe/weight.py
python/freetoken/models/qwen3_moe/__init__.py
python/freetoken/models/qwen3_moe/attention.py
python/freetoken/models/qwen3_moe/config.py
python/freetoken/models/qwen3_moe/model.py
python/freetoken/models/qwen3_moe/moe.py
python/freetoken/models/qwen3_moe/weight.py
python/freetoken/moe/__init__.py
python/freetoken/moe/base.py
python/freetoken/moe/bench_profile.py
python/freetoken/moe/benchbw.py
python/freetoken/moe/cpu_executor.py
python/freetoken/moe/cpu_offload.py
python/freetoken/moe/expert_banks.py
python/freetoken/moe/fused.py
python/freetoken/moe/fused_ds_fp4.py
python/freetoken/moe/fused_fp8_block.py
python/freetoken/moe/fused_mxfp4.py
python/freetoken/moe/fused_nvfp4.py
python/freetoken/moe/fused_q4_0.py
python/freetoken/moe/host_banks.py
python/freetoken/moe/nvfp4_backends.py
python/freetoken/moe/offload.py
python/freetoken/moe/offload_cache.py
python/freetoken/moe/offload_kernels.py
python/freetoken/moe/configs/triton_3_5_1/E=128,N=704,device_name=NVIDIA_H100_80GB_HBM3.json
python/freetoken/scheduler/__init__.py
python/freetoken/scheduler/cache.py
python/freetoken/scheduler/config.py
python/freetoken/scheduler/decode.py
python/freetoken/scheduler/io.py
python/freetoken/scheduler/prefill.py
python/freetoken/scheduler/scheduler.py
python/freetoken/scheduler/status.py
python/freetoken/scheduler/table.py
python/freetoken/scheduler/utils.py
python/freetoken/server/__init__.py
python/freetoken/server/access_log_filter.py
python/freetoken/server/accounting.py
python/freetoken/server/anthropic_api.py
python/freetoken/server/anthropic_models.py
python/freetoken/server/api_models.py
python/freetoken/server/api_server.py
python/freetoken/server/args.py
python/freetoken/server/control_api.py
python/freetoken/server/function_call_parser.py
python/freetoken/server/generation.py
python/freetoken/server/launch.py
python/freetoken/server/model_meta.py
python/freetoken/server/openai_api.py
python/freetoken/server/reasoning_parser.py
python/freetoken/server/request_logger.py
python/freetoken/server/request_ring.py
python/freetoken/server/responses_api.py
python/freetoken/server/stats.py
python/freetoken/server/supervisor.py
python/freetoken/shell/__init__.py
python/freetoken/shell/__main__.py
python/freetoken/shell/client.py
python/freetoken/shell/render.py
python/freetoken/shell/tui.py
python/freetoken/tokenizer/__init__.py
python/freetoken/tokenizer/detokenize.py
python/freetoken/tokenizer/server.py
python/freetoken/tokenizer/tokenize.py
python/freetoken/utils/__init__.py
python/freetoken/utils/arch.py
python/freetoken/utils/hf.py
python/freetoken/utils/logger.py
python/freetoken/utils/misc.py
python/freetoken/utils/mp.py
python/freetoken/utils/progress.py
python/freetoken/utils/registry.py
python/freetoken/utils/torch_utils.py