Commit Graph
20951 Commits
Author SHA1 Message Date
Zexi Li 0fa97910f3 fix(llm): require memory_mb for local_path HAMi devices (#25220)
local_path SKUs cannot estimate VRAM from InstantModel; validate HAMi devices set devices[].memory_mb on create/update and deployment.
2026-07-29 10:21:57 +08:00
wanyaoqi 890f022162 Automated cherry pick of #25206: fix(region,host): host register with correct hostname (#25233)
* fix(region,host): host register with correct hostname

* fix(region): guest migrate reuse create params schedtag
2026-07-28 11:00:07 +08:00
wanyaoqi 285a7ff633 feat(host): support ascend npu hami (#25235) 2026-07-28 10:55:37 +08:00
Zexi Li 9fbf557c31 feat(baremetal): support IPMI RMCP+ cipher suite probe and persist (#25183)
Auto-detect working ipmitool -C suites (0/3/17) and store cipher_suite on host IPMI config for subsequent connections.
2026-07-28 10:26:54 +08:00
屈轩 bb35915542 fix(climc): climc options update (#25227) 2026-07-27 19:41:12 +08:00
wanyaoqi 4375554d4c fix(host): qmp device add convert string to boolean (#25229) 2026-07-27 19:08:08 +08:00
Zexi Li 60556ad18f fix(aiproxy): avoid permanent ai_key blacklist and clarify skip reasons (#25223)
Enter cooldown when health score hits zero so keys can recover, and include concrete skip reasons in resolve errors.
2026-07-27 12:15:43 +08:00
Jian QiuandQiu Jian 6f98d8ae4c feature: ip set support (#25218)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2026-07-24 16:34:34 +08:00
Jian QiuandQiu Jian f5da42d16f fix: vpc distributes static routes (#25211)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2026-07-23 19:11:38 +08:00
Zexi Li a1012f738a feat(llm): support hami (#25215) 2026-07-23 17:08:37 +08:00
Jian QiuandQiu Jian bed62d0a37 fix: baremetal use legacy http client (#25208)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2026-07-23 16:33:03 +08:00
屈轩 917617c8ad fix(region): vendor update (#25204) 2026-07-22 11:19:10 +08:00
Jian QiuandQiu Jian 71873273bd fix: allow empty ip on link-local nic (#25200)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2026-07-22 11:06:13 +08:00
屈轩 eea4982b77 fix(region): vendor update (#25197) 2026-07-21 23:38:07 +08:00
屈轩 5f0cb67c4d fix(mcp-server): optimized mcp desc (#25194) 2026-07-21 16:19:25 +08:00
屈轩 6b813e8a4d fix(mcp-server): support forwarding mcp-server request (#25191) 2026-07-21 13:27:43 +08:00
屈轩 29d7fe6cb0 fix(mcp-server): optimized mcp server options (#25188) 2026-07-20 20:56:55 +08:00
屈轩 21822dc778 fix(mcp-server): mcp use climc struct (#25184) 2026-07-20 17:17:24 +08:00
wanyaoqi 574cac011a feat(region,host,scheduler): isolated device refactor and hami support (#25129) 2026-07-20 13:04:51 +08:00
wanyaoqi 31f19deb84 fix(host-deployer): bond netplan add vlan subinterface config (#25177) 2026-07-20 12:16:07 +08:00
wanyaoqi af1c3e03a6 fix(region): baremetal boot mode decided by selected host on image support both uefi&bios (#25153) 2026-07-20 12:07:11 +08:00
wanyaoqi e8d6564b49 fix(host-deployer): windows disk extend separate partition and filesystem (#25139) 2026-07-20 12:05:53 +08:00
wanyaoqi df27859f1d fix(host): add nest virtualization options (#25124) 2026-07-20 12:03:13 +08:00
wanyaoqi cbaa543430 fix(region): disk change driver support change aio_mode and cache_mode (#25094) 2026-07-20 11:38:59 +08:00
屈轩 8cf702d132 fix(mcp-server): query scope params (#25172) 2026-07-17 00:26:52 +08:00
屈轩 5381986ed8 fix(cloudid): batch detach policy (#25164) 2026-07-16 19:27:50 +08:00
屈轩 41f6bdb7ce fix(region): support cas hypervisor (#25161) 2026-07-16 19:07:58 +08:00
Zexi Li ca4c4e8d46 fix(aiproxy): emit done events before response.completed in stream converter (#25157)
Emit output_item.done, reasoning_summary_part.done, function_call_arguments.done,
and related finalize events for text, reasoning, and tool calls before the
response.completed event to match the Responses API stream contract.
2026-07-16 17:02:02 +08:00
Zexi Li 98d4515548 feat(aiproxy): auto-provision placeholder ai_key for upstream resolve (#25155) 2026-07-15 19:24:43 +08:00
cwz_eikoh e6c6f66760 feat(llm): add llm-router-agent (#25149) 2026-07-15 15:32:45 +08:00
cwz_eikoh f3bc838d41 feat(llm): support llm-bench in llm (#25147) 2026-07-15 15:03:15 +08:00
Zexi Li 5505986263 feat(aiproxy): support streaming responses for visual messages (#25145)
* feat(aiproxy): support streaming responses for visual messages

* refactor(aiproxy): share API log helpers and support input/output token usage

Extract common chatlog record lifecycle across proxy handlers, and fall back
to Responses/Anthropic usage field aliases when filling token counts.
2026-07-15 10:57:37 +08:00
Zexi Li 83f4518355 feat(aiproxy): move visual provider to model columns and support streaming responses (#25143)
Promote visual_provider_id/visual_model_key to dedicated ai_model columns,
force non-stream upstream orchestration for visual paths, and synthesize SSE
chunks for streaming Responses requests that include images.
2026-07-14 18:39:44 +08:00
Zexi Li b4e5fbb0b7 refactor(aiproxy): rename chat log config to API log with S3 fields (#25132)
* refactor(aiproxy): rename chat log config to API log with S3 fields

* feat(aiproxy): add visual extension for Responses image tools

Wire visual model catalog, orchestration, and OpenAI Responses compat so chat/Codex can invoke image generation and editing tools.
2026-07-14 11:52:43 +08:00
wanyaoqi 6e889e9cfa fix(region): reset guest quorum child index on guest shutdown (#25134) 2026-07-14 00:08:59 +08:00
屈轩 c75cfb4ca1 fix(region): split sync and probe worker (#25101) 2026-07-13 13:59:34 +08:00
Jian QiuandQiu Jian 2e28bd7ca7 fix: allow security group cidr have multiple prefixes (#25123)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2026-07-13 13:58:51 +08:00
Zexi Li e73cb9dc43 feat(aiproxy): add Z.AI / Zhipu provider with dual OpenAI and Anthropic API modes (#25130)
Register zhipu as a public OpenAI-compat provider with GLM catalog seeds,
Anthropic passthrough routing, and v4-style base URL path handling.
2026-07-13 11:25:09 +08:00
Zexi Li 0d58ed4ccb feat(aiproxy): add OpenAI Responses API with provider adapters and Codex support (#25121)
Implement /v1/responses endpoint with multi-provider translation (OpenAI, Azure,
Anthropic, Gemini), streaming compat layers, Codex config/tool integration, and
updated functional test docs.
2026-07-10 16:40:42 +08:00
Jian QiuandQiu Jian 5bc80cfb42 fix: allow deleting idp without checking project count (#25109)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2026-07-10 08:35:32 +08:00
屈轩 214a606991 fix(region): gcp network tags (#25120) 2026-07-09 17:11:25 +08:00
屈轩 60bc65da31 fix(yunionconf): batch import tags (#25117) 2026-07-09 16:43:06 +08:00
Jian QiuandQiu Jian ff9268966c fix: host order by sql redundant group by (#25113)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2026-07-09 14:22:38 +08:00
cwz_eikoh e0b568d0ed feat(aiproxy): support log record in aiproxy through minio (#25115) 2026-07-09 13:01:35 +08:00
Jian QiuandQiu Jian 95f4bcf1f0 fix: lb listener rule specific certificates (#25105)
Co-authored-by: Qiu Jian <qiujian@yunionyun.com>
2026-07-08 12:03:36 +08:00
Zexi Li f538dcf2e6 feat(aiproxy): add Anthropic Messages API and migrate functional tests to Go (#25097)
Add /v1/messages handler with Anthropic-to-OpenAI translation, upstream
failover, and probe endpoints. Replace shell-based functional test scripts
with pkg/aiproxy/ft and climc test commands; consolidate documentation.
2026-07-07 20:13:03 +08:00
屈轩 8c87c0e57d fix(region): pve host driver compatibility (#25099) 2026-07-03 15:37:58 +08:00
Zexi Li c91b5bc70c feat(llm): add ModelScope instant model import and refactor model download (#25091)
Add ModelScope hub integration, download driver, and instant model import
flow parallel to HuggingFace. Extract shared HuggingFace download helpers
and reuse them across vLLM/SGLang container drivers.
2026-07-01 20:12:59 +08:00
Zexi Li f1f3c1df2a feat(llm): support local_path scheduling with prefer_hosts and deployment restart (#25085) 2026-07-01 13:11:02 +08:00
wanyaoqi f6e5f71fba fix(host-deployer): mount procfs before deploy guestfs (#25086) 2026-07-01 00:53:31 +08:00