What's Changed

• feat: gpt-5.6 support by @zijiren233 in #612
• fix: preserve channel permission for cyber policy 403 errors by @zijiren233 in #615
• fix(doubao): set reasoning_effort field in doubao reasoning conversion by @yumushang in #616
• fix: prevent Redis rate limiter scripts from blocking by @zijiren233 in #618

New Contributors

• @yumushang made their first contribution in #616

Full Changelog: v0.6.5...v0.6.6

What's Changed

• Revert "fix anthropic streaming content_block_start struct bug" by @zijiren233 in #609

Full Changelog: v0.6.4...v0.6.5

What's Changed

• fix: gemini vertexai audio data need convert http to base64 by @zijiren233 in #608

Full Changelog: v0.6.3...v0.6.4

What's Changed

• fix anthropic streaming content_block_start struct bug by @huangwei1024 in #604
• fix: responses argument type by @zijiren233 in #605

New Contributors

• @huangwei1024 made their first contribution in #604

Full Changelog: v0.6.2...v0.6.3

What's Changed

• feat: respnoses api error cache handler by @zijiren233 in #599
• feat: chat to responses convert filds by @zijiren233 in #600
• fix: chat responses mode finish_reason by @zijiren233 in #601
• feat: chat-responses-reasoning by @zijiren233 in #602

Full Changelog: v0.6.1...v0.6.2

What's Changed

• fix(openai): avoid duplicate responses stream text and normalize mini reasoning by @zijiren233 in #598

Full Changelog: v0.6.0...v0.6.1

What's Changed

• feat: qianfan reasoning_effort convert support by @zijiren233 in #562
• feat: max body size control and max image gen size by @zijiren233 in #563
• chore: update image gen default timeout by @zijiren233 in #564
• feat: ali qwen image support by @zijiren233 in #569
• feat: add more tokens by @zijiren233 in #570
• feat: ali video support by @zijiren233 in #573
• feat: gemini audio video support by @zijiren233 in #574
• feat: siliconflow support audio image video by @zijiren233 in #575
• feat: support doubao image video model by @zijiren233 in #577
• feat: gemini image audio video support by @zijiren233 in #578
• fix: summary price by @zijiren233 in #579
• refactor: gemini/vertexai async usage by @zijiren233 in #584
• fix: async usage consume and disable resolution fuzzy config by @zijiren233 in #585
• fix: allowed_resolutions and fuzzy match allowed_resolutions by @zijiren233 in #587
• feat: add more builtin model by @zijiren233 in #588
• fix: adaptor proto convert and async usage store load metadata by @zijiren233 in #589
• fix: base64 image need clear image url by @zijiren233 in #590
• chore: flattening model columns by @zijiren233 in #591
• fix: gemini image video some error and converts by @zijiren233 in #592
• feat: add ali doubao native video mode by @zijiren233 in #593
• feat: sonic convert and convert error handler by @zijiren233 in #595
• fix: some video native bug and web api doc by @zijiren233 in #594
• feat: add ali qwen models by @zijiren233 in #596
• fix: model config change show by @zijiren233 in #597

Full Changelog: v0.5.8...v0.6.0

What's Changed

• fix: sonic ast cache need clean after middleware by @zijiren233 in #560

Full Changelog: v0.5.7...v0.5.8

What's Changed

• feat: qianfan adaptor support custom appid header by @zijiren233 in #545
• fix: priority help by @zijiren233 in #546
• feat: support qianfan native responses api by @zijiren233 in #547
• feat: support async usage by @zijiren233 in #548
• fix: kimi/moonshot reasoning convert and docs by @zijiren233 in #549
• fix: aws arn modelid need use actual name by @zijiren233 in #550
• feat: jina image embed support by @zijiren233 in #551
• feat: some adaptor support vision embeddings by @zijiren233 in #557
• chore(deps): bump github.com/redis/go-redis/v9 from 9.18.0 to 9.19.0 by @dependabot[bot] in #556
• chore(deps): bump github.com/testcontainers/testcontainers-go from 0.39.0 to 0.42.0 by @dependabot[bot] in #554
• fix: siliconflow embed error handler by @zijiren233 in #558
• chore(deps): bump github.com/larksuite/oapi-sdk-go/v3 from 3.5.3 to 3.6.1 by @dependabot[bot] in #555
• chore(deps): bump github.com/aws/aws-sdk-go-v2/service/bedrockruntime from 1.50.5 to 1.50.6 by @dependabot[bot] in #553
• 调整阿里云百炼对接anthropic接口，原接口只支持qwen3-coder，不支持制定模型 by @mosesyu95 in #528

New Contributors

• @mosesyu95 made their first contribution in #528

Full Changelog: v0.5.6...v0.5.7

What's Changed

• chore: bump go mod and fix mcp panic by @zijiren233 in #525
• feat: support aws claude arn aws bedrock by @zijiren233 in #530
• feat: add openai adaptor reasoning_content map config by @zijiren233 in #532
• fix: mapping glm error by @zijiren233 in #533
• feat: deepseek anthropic native mode support by @zijiren233 in #534
• feat: openai adaptor stream image support by @zijiren233 in #535
• feat: doubao qwen response native mode support by @zijiren233 in #536
• feat: qianfan streamlake baiduv2 support gemini/anthropic by @zijiren233 in #537
• fix: mcp redis cache by @zijiren233 in #538
• feat: adaptor reason effort convert by @zijiren233 in #539
• fix: siliconflow error handloer by @zijiren233 in #540
• feat: config load cache and add fakeerror adaptor by @zijiren233 in #541
• fix: responses mode web search count by @zijiren233 in #542
• fix: store need save origin model name by @zijiren233 in #543

Full Changelog: v0.5.5...v0.5.6