Commit graph

724 commits

Author SHA1 Message Date
Charlie Gleason
3ef72fe8f6
fix(provider): route non-native Cloudflare AI Gateway providers via the REST API (#44828)
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-08-24 22:01:17 -05:00
Charlie Gleason
f8b4dd70ac
fix(provider): send Anthropic's dashed native slug through the AI Gateway (#44281)
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
2026-08-24 16:15:26 -05:00
joelstucki-taulia
3a4c253969
fix(provider): guard textVerbosity injection for @ai-sdk/openai-compatible providers (#43915)
Co-authored-by: Joel Stucki <joel.stucki@example.com>
2026-08-21 22:19:33 +00:00
opencode-agent[bot]
361a71ffad
fix(opencode): route Vertex multi-regions through REP (#42648)
Co-authored-by: Aiden Cline <rekram1-node@users.noreply.github.com>
Co-authored-by: Filip <34747899+neriousy@users.noreply.github.com>
2026-08-21 15:23:05 +02:00
opencode-agent[bot]
71d08e94d5
fix(opencode): retry xAI capacity stream errors (#43640)
Co-authored-by: rekram1-node <rekram1-node@users.noreply.github.com>
2026-08-20 09:29:30 -05:00
opencode-agent[bot]
8b65fa2ef6
fix(opencode): remove Qwen sampling defaults (#43310)
Co-authored-by: rekram1-node <63023139+rekram1-node@users.noreply.github.com>
2026-08-18 17:59:59 -05:00
opencode-agent[bot]
5a0e07efcf chore: generate 2026-08-17 04:54:34 +00:00
Aiden Cline
cba6b5f2f7
feat(opencode): native OpenAI and Anthropic passthroughs for Cloudflare AI Gateway (#42634)
Co-authored-by: Keefe Tang <keefe@pentagram.me>
2026-08-16 23:53:20 -05:00
Matthew Feroz
8571a922db
fix(provider): add Merge Gateway reasoning variants (#41867) 2026-08-12 10:31:19 -05:00
opencode-agent[bot]
5d953482ab
fix(provider): scope DeepSeek V4 Flash sampling defaults (#41620)
Co-authored-by: Aiden Cline <63023139+rekram1-node@users.noreply.github.com>
Co-authored-by: 侯秀凯 <9627034+hoooou@users.noreply.github.com>
2026-08-10 23:01:56 -05:00
Frederik
87481f2bb6
fix: gpt-5.5+ in combination with azure fails with reasoningEffort (#40265) 2026-08-03 17:42:30 -05:00
opencode-agent[bot]
a1ab489e61
fix(provider): widen interleaved reasoning fields (#39556)
Co-authored-by: Aiden Cline <rekram1-node@users.noreply.github.com>
2026-07-30 11:10:34 -05:00
opencode-agent[bot]
5e39051183
fix(provider): omit deprecated Gemini sampling defaults (#38924)
Co-authored-by: Aiden Cline <rekram1-node@users.noreply.github.com>
2026-07-27 23:21:00 -05:00
Aiden Cline
2b2aacc939
fix(provider): generalize Claude adaptive thinking (#38757) 2026-07-24 16:44:00 -05:00
Aiden Cline
542ba88602
fix(provider): select prompt cache keys by SDK (#38424) 2026-07-22 22:37:22 -05:00
Aiden Cline
50eee1f5a4
fix(provider): correct MiniMax M3 thinking variants (#38330) 2026-07-22 10:38:23 -05:00
Aiden Cline
3033afba51
fix(provider): normalize Mistral family tool call IDs (#37982)
Co-authored-by: tobwen <1864057+tobwen@users.noreply.github.com>
2026-07-20 13:58:49 -05:00
Matthias Reso
5a8ee27254
fix(provider): update muse-spark reasoning default and system prompt (#37562)
Co-authored-by: Aiden Cline <aidenpcline@gmail.com>
2026-07-20 13:15:42 -05:00
opencode-agent[bot]
3b719aba48 chore: generate 2026-07-20 04:55:12 +00:00
Qiping Li
20a3a2138e
feat(opencode): use adaptive thinking effort for kimi family on anthr… (#37696)
Co-authored-by: liqiping <liqiping@msh.team>
Co-authored-by: Aiden Cline <aidenpcline@gmail.com>
2026-07-19 23:54:00 -05:00
opencode-agent[bot]
67caf894e0
fix(opencode): increase OpenAI header timeout (#37770)
Co-authored-by: Aiden Cline <rekram1-node@users.noreply.github.com>
2026-07-19 10:41:48 -05:00
Aiden Cline
49d2dd8a38
fix(provider): honor reasoning option semantics (#37519) 2026-07-17 14:23:33 -05:00
Aiden Cline
95ebf50ace
fix(provider): restore Azure Cognitive Services endpoints (#37340) 2026-07-16 12:58:02 -05:00
opencode-agent[bot]
b6478dcebf
fix(opencode): default Meta reasoning to xhigh (#36976)
Co-authored-by: Aiden Cline <rekram1-node@users.noreply.github.com>
2026-07-14 19:52:04 -05:00
Aiden Cline
e434ce01d3
fix(provider): support OpenAI pro reasoning mode (#36694) 2026-07-13 11:04:46 -05:00
Mark
49d997aec3
fix(xai): default store to false for Responses (#36629)
Co-authored-by: Aiden Cline <aidenpcline@gmail.com>
2026-07-13 10:37:12 -05:00
Aiden Cline
99668cfdce
fix(provider): tweak reasoning option driven max budget variants (#36626) 2026-07-13 09:18:32 -05:00
Aiden Cline
a8062ea314
fix(provider): derive variants from reasoning metadata (#36624) 2026-07-13 01:21:59 -05:00
Aiden Cline
8168f0f0f6
fix(provider): route gateway variants by api id (#36614) 2026-07-13 00:08:08 -05:00
Aiden Cline
6f8e1dda15
fix(provider): derive variants from reasoning options (#36543) 2026-07-12 16:12:07 -05:00
Aiden Cline
b8374b5a7c
fix(copilot): handle zero billing batch size (#36123) 2026-07-09 13:14:33 -05:00
Aiden Cline
2e43d41af0
tweak: adjust variant logic for muse (#36079) 2026-07-09 09:07:02 -05:00
Aiden Cline
1db5c2402c
fix(opencode): expose Grok reasoning variants (#36017) 2026-07-09 00:13:10 -05:00
opencode-agent[bot]
9644a008a4 chore: generate 2026-07-08 22:31:09 +00:00
Aiden Cline
ccb6b7c3ea
fix: improve xai cache hit rate (#35970) 2026-07-08 17:29:36 -05:00
opencode-agent[bot]
68f225a11d
fix(provider): preserve OpenRouter small model effort (#35478)
Co-authored-by: Aiden Cline <63023139+rekram1-node@users.noreply.github.com>
2026-07-05 18:50:25 -05:00
Aiden Cline
373cd08b98
fix(copilot): honor advertised model endpoints (#34958) 2026-07-02 10:23:57 -05:00
Aiden Cline
55552c521f
fix(provider): force openai reasoning variants (#34702) 2026-06-30 22:45:39 -05:00
Aiden Cline
f1407e41c4
fix(core): stop replaying stale GitHub Copilot Responses item IDs (#34686) 2026-06-30 19:19:02 -05:00
Aiden Cline
3a669d528b
fix(provider): enable sonnet 5 adaptive thinking (#34673) 2026-06-30 14:03:05 -05:00
James Long
91561bbdfb
refactor(opencode): remove app service layer exports (#34517) 2026-06-29 23:03:57 -04:00
opencode-agent[bot]
9903abc704
fix(opencode): allow empty provider config default (#34167)
Co-authored-by: Aiden Cline <63023139+rekram1-node@users.noreply.github.com>
Co-authored-by: opencode-agent[bot] <opencode-agent[bot]@users.noreply.github.com>
2026-06-26 23:19:42 -05:00
Aiden Cline
ded29f03f0
fix(core): refine small model defaults (#33926) 2026-06-25 20:31:53 -05:00
James Long
0e731c2a1a
refactor(core): support tiered layer nodes (#33937) 2026-06-25 14:34:17 -04:00
Kit Langton
7854f5b9f7
refactor(schema): normalize module patterns (#33770) 2026-06-25 15:42:22 +00:00
Aiden Cline
c17b9557f1
fix(core): preserve structured error messages (#33530) 2026-06-23 18:43:59 -05:00
Kit Langton
4f1ae1604e
chore: upgrade Effect to beta 83 (#32340) 2026-06-20 19:57:29 -04:00
imranshaiedi-byte
22cc758b1a
feat(opencode): expose High/Max thinking variants for GLM-5.2 (#32446)
Co-authored-by: Aiden Cline <aidenpcline@gmail.com>
2026-06-20 16:59:08 -05:00
卫斯李
babe5070e2
fix(opencode): use toLowerCase for Devstral model detection (#33109)
Co-authored-by: Robin1987China <fanrenzhige@163.com>
2026-06-20 14:19:27 -05:00
Jason Quense
213ff3f2d7
fix(opencode): sanitize OpenAI MCP tool schemas (#32489)
Co-authored-by: jquense <jquense@ramp.com>
Co-authored-by: Aiden Cline <aidenpcline@gmail.com>
Co-authored-by: Aiden Cline <63023139+rekram1-node@users.noreply.github.com>
2026-06-17 02:00:58 +02:00