Artem PerevedentsevandGitHub b92ef9ec5a [Perf] Enable FlashInfer top-k/top-p sampler by default (#40376) ...

Signed-off-by: Artem Perevedentsev <aperevedents@nvidia.com>

2026-04-29 19:10:34 +04:00

..

__init__.py

[V1] Adding min tokens/repetition/presence/frequence penalties to V1 sampler (#10681)

2024-12-26 19:02:58 +09:00

test_batched_count_greater_than.py

[Performance Improvement] Update batched_count_greater_than to handle batch size 1 without recompile (#38933)

2026-04-09 23:51:31 +08:00

test_logprobs_e2e.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00

test_logprobs.py

[Bugfix] Fix V1 logprobs empty strings for multi-byte UTF-8 tokens when logprobs > 0 (#34875)

2026-04-08 15:30:00 +00:00

test_rejection_sampler.py

[Feat] Unified Synthetic Acceptance Rate for V1 and V2 (#40662)

2026-04-24 00:48:08 +00:00

test_sampler.py

refactor hard coded device string in test files under tests/v1 and tests/lora (#37566)

2026-04-03 11:21:47 +08:00

test_sampling_params_e2e.py

[V0 Deprecation] Remove code related to per-request logits processors (#34400)

2026-02-12 20:44:28 +08:00

test_topk_topp_sampler.py

[Perf] Enable FlashInfer top-k/top-p sampler by default (#40376)

2026-04-29 19:10:34 +04:00

utils.py

[Reasoning] Support for speculative decoding with thinking budget (#34668)

2026-04-29 06:14:52 +00:00