v0.0.10

版本发布时间: 2023-09-26 20:52:32

InternLM/lmdeploy最新发布版本:v0.6.0a0(2024-08-26 17:12:19)

What's Changed

[feature] Graceful termination of background threads in LlamaV2 by @akhoroshev in https://github.com/InternLM/lmdeploy/pull/458
expose stop words and filter eoa by @AllentDan in https://github.com/InternLM/lmdeploy/pull/352

Fix side effect brought by supporting codellama: sequence_start is always true when calling model.get_prompt by @lvhan028 in https://github.com/InternLM/lmdeploy/pull/466
Miss meta instruction of internlm-chat model by @lvhan028 in https://github.com/InternLM/lmdeploy/pull/470
[bug] Fix race condition by @akhoroshev in https://github.com/InternLM/lmdeploy/pull/460
Fix compatibility issues with Pydantic 2 by @aisensiy in https://github.com/InternLM/lmdeploy/pull/465
fix benchmark serving cannot use Qwen tokenizer by @AllentDan in https://github.com/InternLM/lmdeploy/pull/443
Fix memory leak by @lvhan028 in https://github.com/InternLM/lmdeploy/pull/488

@eltociear made their first contribution in https://github.com/InternLM/lmdeploy/pull/462
@akhoroshev made their first contribution in https://github.com/InternLM/lmdeploy/pull/458
@aisensiy made their first contribution in https://github.com/InternLM/lmdeploy/pull/465

Full Changelog: https://github.com/InternLM/lmdeploy/compare/v0.0.9...v0.0.10