What's Changed

🐛 Bug Fixes

• fix(distribution): honor proxies in the guarded token-exchange client (#1040) @ignasi

This release fixes a critical issue where model pulls from authenticated registries (including Docker Hub) would fail in proxied deployments. The SSRF guard was incorrectly rejecting proxy connections on private or loopback addresses, preventing token exchange operations from completing successfully.

Full Changelog: v1.2.7...v1.2.8

What's Changed

🚀 Features

• Add structured output support to Responses API (#1010) @dmedovich

🐛 Bug Fixes

• fix(distribution): validate token realm on pull and re-challenge paths (#1038) @ignasi
• fix(registry): honor credential store when fetching backend images (#1033) @doringeman

📦 Dependencies

• chore(deps): bump the github-actions group with 2 updates (#1027) @dependabot
• chore(deps): bump brace-expansion (#1021) @dependabot
• chore: bump llama.cpp (b10068) (#1019) @github-actions
• chore(deps): bump the github-actions group with 7 updates (#1013) @dependabot
• chore: bump llama.cpp (b9982) (#1011) @github-actions

🔧 Maintenance

• ci(release): guard run-ID polling against transient gh failures (#1007) @ignasi
• Build dmr.exe on Windows in build-dmr target (#1006) @eric-curtin
• ci(release): make downstream release triggers idempotent and race-free (#1005) @ignasi

Full Changelog: v1.2.6...v1.2.7

What's Changed

🚀 Features

• Add an optional CommandModifier for backend runner processes (#1003) @mat007
• Pin llama.cpp version per release, allow pulling latest on demand, and make llama.cpp a deferred backend on macOS/Windows (#737) @doringeman
• Rewrite dmr as a standalone, engine-independent binary (#996) @ericcurtin
• Show size of model when using docker model search (#988) @KeeTraxx
• Add helper + tests for SyncDockerConfigToContainer (#947) @areebahmeddd

🐛 Bug Fixes

• Add Pascal GPU architecture support to llama.cpp build (#986) @lohitkolluri

📦 Dependencies

• Bump llama.cpp version for macOS and Windows (#1002) @docker
• Bump llama.cpp for Linux (#1001) @doringeman
• Bump llama.cpp to b9879 (#997) @docker
• Bump llamacpp to b9859 (#995) @krissetto
• Bump llama.cpp to b9837 (#990) @docker

🔧 Maintenance

• Pin actions to commit SHAs, drop 3p release action (#1000) @docker
• Update configure show command description and documentation (#928) @Montana

Full Changelog: v1.2.5...v1.2.6

What's Changed

• chore(deps): bump actions/checkout from 6.0.3 to 7.0.0 in the github-actions group by @dependabot[bot] in #991
• fix: use existing formatSize function, size regards model files only by @KeeTraxx in #988
• fix(cuda): add Pascal GPU architecture support to llama.cpp build by @lohitkolluri in #986
• chore: bump llama.cpp (https://github.com/ggml-org/llama.cpp/releases/b9837) by @github-actions[bot] in #990
• cli: hint when install-runner options are ignored by @Yash-2003P in #938
• chore(deps): bump multer from 2.1.1 to 2.2.0 in /demos/extractor in the npm_and_yarn group across 1 directory by @dependabot[bot] in #977
• sync docker credentials to container by @areebahmeddd in #947
• docs: clarify configure show command help by @Montana in #928
• Bump llamacpp to b9859 by @krissetto in #995
• chore(deps): bump the github-actions group with 3 updates by @dependabot[bot] in #999
• Rewrite dmr as a standalone, engine-independent binary by @ericcurtin in #996
• chore: bump llama.cpp (https://github.com/ggml-org/llama.cpp/releases/b9879) by @github-actions[bot] in #997

New Contributors

• @lohitkolluri made their first contribution in #986
• @Yash-2003P made their first contribution in #938
• @Montana made their first contribution in #928

Full Changelog: v1.2.5...dmr-v0.1.0

What's Changed

🚀 Features

• Show size of model when using docker model search (#985) @KeeTraxx

🐛 Bug Fixes

• fix(registry): preserve mirror URL path prefix when resolving/pulling backend images (#987)
• Override healthcheck in dockerfile (#946) @areebahmeddd

📦 Dependencies

• chore(deps): bump github.com/containerd/containerd/v2 (#982) @dependabot
• chore(deps): bump docker/cagent-action in the github-actions group (#981) @dependabot

🔧 Maintenance

• Migrate cagent-action to docker-agent-action (v2.0.0) (#984)

Full Changelog: v1.2.4...v1.2.5

What's Changed

🚀 Features

• feat: update speculative decoding flags for llama.cpp v1.5+ (#956) @HelloItMeMort

🐛 Bug Fixes

• fix(llamacpp): resolve backend tag via registry API to honor mirrors (#974) @ignLopez

📦 Dependencies

• chore: bump diffusers to 0.38.0 (#975) @ignLopez
• chore(deps): bump actions/checkout in the github-actions group (#973) @dependabot
• chore: bump llama.cpp to b9592 @ignLopez

🔧 Maintenance

• ci(release): prune cloud builder cache before building (#968) @ignLopez

Full Changelog: v1.2.3...v1.2.4

What's Changed

📦 Dependencies

• chore: bump llama.cpp to b9592 (#967) @ilopezluna

Full Changelog: v1.2.2...v1.2.3

What's Changed

🚀 Features

• Add Open WebUI volume mount and format containerApp entry (#944) @chyipin

🐛 Bug Fixes

• Fix install libllama-server-impl.dylib to install/lib on macOS (#945) @ilopezluna
• Make llama.cpp Python cleanup optional (#957) @ilopezluna

📦 Dependencies

• Bump llama.cpp to b9501 (#955) @ilopezluna
• Bump llama.cpp to b9309 (#937) @github-actions
• Bump the github-actions group with 5 updates (#939) @dependabot
• Bump go.opentelemetry.io/otel/sdk from 1.41.0 to 1.43.0 (#935) @dependabot
• Bump github.com/containerd/containerd/v2 to 2.2.4 (#934) @dependabot
• Bump the github-actions group with 2 updates (#950) @dependabot

🔧 Maintenance

• Add Dependabot cooldown (#953) @securityeng-bot
• Add npm lockfile for demos/extractor (#926) @docker-agent

Full Changelog: v1.2.1...v1.2.2

What's Changed

🚀 Features

• Honor registry mirrors for model pulls and backend downloads (#931) @doringeman
• Add vLLM ROCm Docker variant for AMD GPUs (#917) @dmedovich
• Add MUSA and OpenVINO support using upstream llama.cpp images (#897)
• Run daemon and client from a single binary (#900) @areebahmeddd
• Use official llama.cpp images for Linux (#879)
• Add ContextSize path from FromDirectory (#895) @sathira10
• Extract context size (#866) @sathira10

🐛 Bug Fixes

• Use rm -f to handle missing .py files in llamacpp install (#932)
• Fix gosec g602 error for make validate-all (#911) @areebahmeddd
• Use path.Join instead of filepath.Join for tar entry names (#898)
• Increase timeout for E2E tests to improve stability (#891)
• Update release workflow to prioritize GitHub Release tags (#890)
• Optimize latest release tag retrieval in release.yml (#896) @IgnasiBarrera
• Cap max_tokens in E2E chat helpers to prevent test timeouts (#893)
• Add path sanitization to prevent directory traversal in downloader (#889) @IgnasiBarrera
• Pin Python to 3.12 in vLLM and SGLang build stages (#883)

📦 Dependencies

• Bump llama.cpp to b9209 (#919)
• Bump llama.cpp to b9174 (#918) @krissetto
• Bump llama.cpp to b9102 (#908)
• Bump llama.cpp to b9014 (#901)
• Bump llama.cpp to b8943 (#885)
• Bump docker/cagent-action from 1.5.1 to 1.5.3 (#920)
• Bump docker/cagent-action in the github-actions group (#909)
• Bump openssl (#904)
• Bump the go-modules-root group with 4 updates (#886)
• Bump rustls-webpki (#888)
• Bump the github-actions group with 2 updates (#887)
• Bump openssl (#882)

🔧 Maintenance

• Remove fragile bump-pinata and update-docs jobs from release workflow (#933) @IgnasiBarrera
• Install vLLM ROCm from upstream wheels (#923) @dmedovich
• Update vllm install to use --torch-backend auto (#921)
• Pin golangci-lint version in .versions and read it in CI (#913) @doringeman
• Declare contents: read on ci/e2e-test/integration-test workflows (#914) @arpitjain799
• Bump model-runner Go dependency in pinata during release (#881) @IgnasiBarrera

Full Changelog: v1.1.38...v1.2.1

What's Changed

🐛 Bug Fixes

• fix: pass chat template to vllm-metal backend (#880) @ilopezluna
• fix: install llama-common DLL on Windows shared builds (#878) @ilopezluna

Full Changelog: v1.1.37...v1.1.38