Awni HannunandGitHub ac8ae2c05a Improve reasoning and tool call parsing in server (#711) ...

* Parse reasoning in server

* redesign and start to fix tool parsing

* add function gemma

* fix

* fix

* glm47 tools

* add minimax m2 tool parser

* tool_call finish reason

* Keep model wired in the server to reduce ttft

* infer tool parser

2026-01-05 14:18:52 -08:00

..

test_chat.py

Add system prompt to chat script (#334)

2025-07-30 08:33:33 -07:00

test_datsets.py

Use test data zipfile in CI (#662)

2025-12-08 16:13:34 -08:00

test_evaluate.py

Batch generation (#443)

2025-09-15 16:02:45 -07:00

test_finetune.py

DWQ for very large models (#536)

2025-11-07 06:43:40 -08:00

test_generate.py

Add logits_processors support to batch_generate (#635)

2026-01-04 15:19:00 -08:00

test_gguf.py

version

2025-03-13 15:37:13 -07:00

test_losses.py

fix flaky test (#643)

2025-12-02 13:33:11 -08:00

test_models.py

Add K-EXAONE MoE (#719)

2026-01-05 08:50:33 -08:00

test_prompt_cache.py

Use test data zipfile in CI (#662)

2025-12-08 16:13:34 -08:00

test_sample_utils.py

Adding support for XTC (eXclude Top Choice) (#110)

2025-04-24 14:25:30 -07:00

test_server.py

Fix empty /v1/models response for locally loaded models (#713)

2025-12-31 14:22:00 -08:00

test_tokenizers.py

DWQ for very large models (#536)

2025-11-07 06:43:40 -08:00

test_tool_parsing.py

Improve reasoning and tool call parsing in server (#711)

2026-01-05 14:18:52 -08:00

test_tuner_trainer.py

DWQ for very large models (#536)

2025-11-07 06:43:40 -08:00

test_tuner_utils.py

version

2025-03-13 15:37:13 -07:00

test_utils.py

Cleanup and simplify model I/O (#532)

2025-10-07 06:57:51 -07:00