huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-10-16 10:25:23 +00:00
Code Issues Packages Projects Releases Wiki Activity
efb94e0d3d
text-generation-inference/server
History
Daniël de Kok c6071749db
Fix mask passed to flashinfer (#3324) 
Custom masks are padded to the shape `[batch_size, max_len, max_len]`.
However, flashinfer expects an unpadded mask of the shape
`[sum(q_len[i] * k_len[i] for i in range(batch_size)]`.

This change unpads the custom mask (currently only used by Gemma 3)
to this shape (assuming q_len == k_len, since we only use the custom
mask during prefill).
2025-09-08 13:47:03 -04:00
..
custom_kernels All integration tests back everywhere (too many failed CI). (#2428) 2024-08-16 21:19:46 +02:00
exllama_kernels Update ROCM libs and improvements (#2579) 2024-09-30 10:54:32 +02:00
exllamav2_kernels Update ROCM libs and improvements (#2579) 2024-09-30 10:54:32 +02:00
tests Small test and typing fixes (#3078) 2025-03-10 15:08:23 +01:00
text_generation_server Fix mask passed to flashinfer (#3324) 2025-09-08 13:47:03 -04:00
.gitignore Impl simple mamba model (#1480) 2024-02-08 10:19:45 +01:00
bounds-from-nix.py Sync (most) server dependencies with Nix (#2782) 2024-12-03 04:04:06 +01:00
kernels.lock Switch to punica-sgmv kernel from the Hub (#3236) 2025-05-21 15:44:15 +02:00
Makefile Switch to punica-sgmv kernel from the Hub (#3236) 2025-05-21 15:44:15 +02:00
Makefile-awq chore: add pre-commit (#1569) 2024-02-16 11:58:58 +01:00
Makefile-eetq Putting back the NCCL forced upgrade. (#2999) 2025-02-14 11:31:59 +01:00
Makefile-exllamav2 Upgrading exl2. (#2415) 2024-08-14 11:58:08 +02:00
Makefile-flash-att Putting back the NCCL forced upgrade. (#2999) 2025-02-14 11:31:59 +01:00
Makefile-flash-att-v2 Add Flash decoding kernel ROCm (#2855) 2025-01-13 11:12:35 +01:00
Makefile-flashinfer flashinfer 0.2.0.post1 -> post2 (#3040) 2025-02-20 12:34:20 +01:00
Makefile-selective-scan chore: add pre-commit (#1569) 2024-02-16 11:58:58 +01:00
Makefile-vllm Use ROCM 6.3.1 (#3141) 2025-04-07 12:55:11 +02:00
pyproject.toml Fix outline import issue (#3282) 2025-08-21 09:53:04 +02:00
README.md chore: add pre-commit (#1569) 2024-02-16 11:58:58 +01:00
req.txt Using the "lockfile". (#2992) 2025-02-06 12:28:24 +01:00
requirements_cuda.txt Update to Torch 2.7.0 (#3221) 2025-05-15 11:48:33 +02:00
requirements_gen.txt Update to Torch 2.7.0 (#3221) 2025-05-15 11:48:33 +02:00
requirements_intel.txt Update to Torch 2.7.0 (#3221) 2025-05-15 11:48:33 +02:00
requirements_rocm.txt Update to Torch 2.7.0 (#3221) 2025-05-15 11:48:33 +02:00
uv.lock chore: prepare version 3.3.5 (#3314) 2025-09-02 15:35:42 +02:00

README.md

Text Generation Inference Python gRPC Server

A Python gRPC server for Text Generation Inference

Install

make install

Run

make run-dev