Release Notes
v26.11.2
On this page
Released 2026-11-08 · Release line 26.11 · Commit 3cf500e60
This release makes GPUWorks easier to use in offline and private-network environments, and fixes reasoning output in the Playground and benchmark metrics, agent connection settings, and API key revocation. All 26.11 users should upgrade, especially those deployed on private networks or offline. This release has no database migration.
Upgrade Notes
- Private-network sources are allowed by default: HTTP sources on ordinary private and loopback addresses, reached from the management node, change from denied by default to allowed by default.
To keep the old behavior, add
outbound_restrict_private_hosts: truetoserver.yamlbefore upgrading; existingoutbound_allow_private_hostsentries keep working. Metadata, link-local, and special reserved addresses stay denied in both modes. - Storage paths:
storage_probe_rootsno longer limits the paths of new or edited storage; it only drives a startup warning about existing storage. - Upgrade the server first, then the agents: the benchmark metric fixes (when the agent generates the load), the Docker startup wait, and
runveil loadrun on the agent, so agents also need v26.11.2.
Added
gpuworks-agent runveil load -i FILE REF: load adocker saveor OCI archive (.tar/.tar.gz) onto a runveil (udocker) node, without the management node reaching an HTTP source and without stopping the agent. See Offline Deployment.server.yamlgainsoutbound_restrict_private_hosts(defaultfalse); when enabled, only private targets granted inoutbound_allow_private_hostsare allowed.server.yamlgainsstorage_denied_paths: replaces the built-in list of directories storage may not use;[]denies none. See Configuration and Environment Variables.- Benchmark reports add time to first visible answer and a count of requests that produced output but no answer text; a streamed request whose usage reports output tokens
but carries no recognizable text raises the warning
BENCHMARK_STREAM_OUTPUT_UNRECOGNIZED. See Run Benchmarks and Read the Report.
Changed
- Storage paths no longer need to sit under
storage_probe_roots(default/opt/gpuworks/models,/opt/gpuworks/cache,/mnt/gpuworks); paths such as/data/modelswork directly. System directories are still denied. - Changes to
outbound_allow_private_hostsandoutbound_restrict_private_hoststake effect on the management node's next outbound request, without restarting the server. - A source denied by the outbound policy is reported as
IMAGE_ARCHIVE_URL_POLICY_DENIEDinstead of "unreachable". - When the Docker daemon is not ready at agent startup, the agent waits up to 2 minutes; if Docker still does not answer, it falls back to runveil with the reason
daemon_not_readyand switches back once Docker is ready. Newly generated systemd units start afterdocker.serviceandcontainerd.service.
Fixed
- A model registered from a node's local directory, paired with an inference template that declares public weight sources such as Hugging Face or ModelScope, kept failing evaluation with a source conflict and could not be deployed. (GWKS-216)
x.part/x.part.statefiles left by a cancelled platform download were still treated as an unfinished download after another tool finishedx, so the model could not be registered or scanned.- Installing or restarting a nohup agent inherited stale
GPUWORKS_CONNECT_MODE,GPUWORKS_SERVER,GPUWORKS_TOKEN, and similar variables from the shell, reverting the connection settings already written to the configuration. - An agent that started before Docker fell back to runveil permanently, and multi-node deployments were refused because runveil cannot expose RDMA devices.
- Newer vLLM versions stream reasoning under
delta.reasoning(so does thecuda12.9-vllm0.31.0-r2runner shipped with 26.11), and the Playground and AI assistant showed no reasoning. They now readreasoning_content,reasoning, andthinking. - The benchmark load generator did not read
delta.reasoning: TTFT absorbed the whole reasoning time, TPOT came out too low, and requests that spent their output budget reasoning were left out of the TTFT/TPOT samples.
Security
- Revoking an API key prefix that does not exist returns 404
API_KEY_NOT_FOUNDinstead of reporting it revoked; an authentication already in flight before a revocation cannot admit new requests after it. (GWKS-203) klauspost/compressmoves to 1.20.2 for GO-2026-6865; the library decompresses imported OCI image layers and downloaded archives. (GWKS-214)- Built with Go 1.26.10 for GO-2026-6809 in
crypto/x509, reachable through server_dial.
Downloads
Archives live under https://dl.gpuworks.ai/releases/v26.11.2/; the same paths on Alibaba Cloud OSS and Huawei Cloud OBS hold identical files.
| File | SHA-256 |
|---|---|
gpuworks-v26.11.2-linux-amd64.tar.gz | 6b1762f0047819b12dd1750a514e8d38f4a602ca2efa2445b31d412f510edc66 |
gpuworks-v26.11.2-linux-arm64.tar.gz | 1498dce13f019b106e35adb2ab701197430372af6d6aa29b9071e841b937af74 |
gpuworks-v26.11.2-darwin-arm64.tar.gz | 00e5ef7e08cea99f26d20644ef6e1b219aab0a83f49ad45f3654118aba8a6cb8 |
gpuworks-v26.11.2-darwin-amd64.tar.gz | be2ce55b2e1cd61aa171ef21fb752c34326b50de4e8cc5747ec795fc3fdfd60e |
Full Changelog
Was this page helpful?