v2.7.0

# What's new in 2.7.0 (2026-04-25)
These are the changes in inference v2.7.0.
## New features
* FEAT: support replica removing by @leslie2046 in https://github.com/xorbitsai/inference/pull/4784
* feat(ui): optimize configuration cache dialog and terminology by @leslie2046 in https://github.com/xorbitsai/inference/pull/4807
* FEAT: add DeepSeek V3.2 tool parser for DSML format by @amumu96 in https://github.com/xorbitsai/inference/pull/4771
* feat: add Glm4MoeLiteForCausalLM support by @amumu96 in https://github.com/xorbitsai/inference/pull/4835
* feat: parallelize multi-replica terminate_model and improve UI delete UX by @m199369309 in https://github.com/xorbitsai/inference/pull/4825
* FEAT: [model] qwen3.6 support by @llyycchhee in https://github.com/xorbitsai/inference/pull/4831
* FEAT: [model] MiniMax-M2.7 support by @llyycchhee in https://github.com/xorbitsai/inference/pull/4843
* feat(tool_parser): add plain format support to DeepSeek V3.2 tool parser by @amumu96 in https://github.com/xorbitsai/inference/pull/4842
* FEAT: [model] glm-5.1 support by @llyycchhee in https://github.com/xorbitsai/inference/pull/4832
## Enhancements
* ENH: update model "qwen3.5" JSON by @llyycchhee in https://github.com/xorbitsai/inference/pull/4801
* ENH: update model "DeepSeek-V3.2" JSON by @amumu96 in https://github.com/xorbitsai/inference/pull/4813
* ENH: update models JSON [embedding] by @XprobeBot in https://github.com/xorbitsai/inference/pull/4824
* ENH: update model "qwen3.5" JSON by @qinxuye in https://github.com/xorbitsai/inference/pull/4821
* ENH: update models JSON [embedding, rerank] by @XprobeBot in https://github.com/xorbitsai/inference/pull/4841
* ENH: update 2 models JSON ("MiniMax-M2.7", "glm-5.1") by @XprobeBot in https://github.com/xorbitsai/inference/pull/4848
* BLD: update xllamacpp to newest version for docker by @qinxuye in https://github.com/xorbitsai/inference/pull/4819
* BLD: Remove pre-release PyTorch installation from Dockerfile by @zwt-1234 in https://github.com/xorbitsai/inference/pull/4836
* BLD: remove torch related installation in aarch64 dockerfile by @zwt-1234 in https://github.com/xorbitsai/inference/pull/4840
* BLD: remove torchcodec installation in Dockerfile aarch64 by @zwt-1234 in https://github.com/xorbitsai/inference/pull/4844
* BLD: Modify docker build command for aarch64 image by @zwt-1234 in https://github.com/xorbitsai/inference/pull/4853
## Bug fixes
* fix: replace eval() with safe alternatives to prevent RCE in tool parsers by @Ricardo-M-L in https://github.com/xorbitsai/inference/pull/4786
* fix: support JSON object parameters in CLI by @Ricardo-M-L in https://github.com/xorbitsai/inference/pull/4787
* fix: support Jina API task parameters for jina-embeddings-v4 by @Ricardo-M-L in https://github.com/xorbitsai/inference/pull/4788
* fix(ui): handle mixed dict and ChatMessage types in history by @qinxuye in https://github.com/xorbitsai/inference/pull/4814
* fix(vllm): fix gemma-4 tool calls by @llyycchhee in https://github.com/xorbitsai/inference/pull/4815
* fix(docker): unpin torchcodec to fix 503 error on reranker/embedding model load by @FlintyLemming in https://github.com/xorbitsai/inference/pull/4817
* fix: handle missing 'cpu' key in get_cluster_device_info to prevent KeyError 500 by @m199369309 in https://github.com/xorbitsai/inference/pull/4822
* fix: venv concurrent creation race, cold-start lock dir, and jina-embeddings-v4 torch mismatch by @m199369309 in https://github.com/xorbitsai/inference/pull/4823
* fix: dynamic CUDA version check for extra_index_url by @Gmgge in https://github.com/xorbitsai/inference/pull/4820
* fix: vLLM multi-node distributed init and pipeline parallel inference by @amumu96 in https://github.com/xorbitsai/inference/pull/4834
* fix: venv torchvision alignment, supervisor RPC timeouts, get_model flood protection, replica pre-check, and safe log handler by @m199369309 in https://github.com/xorbitsai/inference/pull/4839
* fix: remove last message role restriction in chat completion endpoint by @amumu96 in https://github.com/xorbitsai/inference/pull/4833
* fix(security): prevent pwn-request vulnerability in gen_docs workflow by @qinxuye in https://github.com/xorbitsai/inference/pull/4850
## Documentation
* DOC: add v2.5.0 release notes by @qinxuye in https://github.com/xorbitsai/inference/pull/4800
## Others
* refactor(device_utils): replace if/elif chains with DeviceSpec registry by @amumu96 in https://github.com/xorbitsai/inference/pull/4846

## New Contributors
* @Ricardo-M-L made their first contribution in https://github.com/xorbitsai/inference/pull/4786
* @FlintyLemming made their first contribution in https://github.com/xorbitsai/inference/pull/4817
* @m199369309 made their first contribution in https://github.com/xorbitsai/inference/pull/4822

**Full Changelog**: https://github.com/xorbitsai/inference/compare/v2.5.0...v2.7.0