8fb1e877cd
fix(ai): disable hidden provider 429 retries ( #4991 )
2026-05-26 23:41:58 +02:00
71446c6c2b
fix(ai): correct Codex Spark context window
...
closes #4969
2026-05-26 09:08:40 +02:00
fa1180b6b7
fix(ai): detect Poolside context overflow
...
Closes #4943
2026-05-26 08:58:21 +02:00
fc8a155901
fix(ai): honor Codex Responses maxRetries
2026-05-25 12:30:04 +02:00
83a227aed3
Update release instructions and generated models
2026-05-23 12:02:35 +02:00
d801d88a11
Support adaptive thinking for Anthropic-compatible aliases
...
closes #4790
2026-05-22 18:30:51 +02:00
c841a6c78f
Clean up OAuth device-code callbacks
2026-05-22 15:51:36 +02:00
11e868b790
Merge pull request #4788 from earendil-works/refactor-device-code-login
...
feat(ai): refactor device code login for copilot
2026-05-22 15:21:42 +02:00
1a2a536dba
chore: update PR prompt template
2026-05-22 11:11:38 +02:00
c554364c2a
feat(ai): refactor device code login for copilot
2026-05-22 08:54:18 +02:00
11c3da4f74
fix(ai): set bedrock claude default max tokens
...
closes #4848
2026-05-21 23:27:40 +02:00
b141e1fa24
chore: remove web-ui workspace
2026-05-20 02:26:09 +02:00
ae9450dc51
chore(ts): use source import extensions
2026-05-20 00:04:03 +02:00
06c6c324d7
chore: enforce erasable TypeScript syntax
2026-05-19 23:15:39 +02:00
2787b601d7
fix(ai): stop defaulting max token request caps
...
closes #4675
2026-05-19 11:49:44 +02:00
7be75baded
fix(ai): clamp OpenAI prompt cache keys
...
closes #4720
2026-05-19 10:51:11 +02:00
0c4d704a7b
Add [Unreleased] section for next cycle
2026-05-18 11:47:16 +02:00
b8f51957a0
fix(ai): add Xiaomi reasoning replay compat
...
closes #4678
2026-05-18 11:15:20 +02:00
9e2bfc7c40
fix(ai): skip unknown bedrock content blocks
...
closes #4223
2026-05-18 01:42:18 +02:00
52e13870a1
fix(ai): prefix HTTP status codes onto Azure OpenAI and OpenAI Responses error messages so auto-retry fires on 5xx/429
...
closes #4232
2026-05-18 01:11:23 +02:00
21d80deda2
fix(ai): normalize opencode go reasoning replay
...
closes #4251
2026-05-18 01:11:23 +02:00
ed3904ddd3
fix(ai): switch xiaomi models to openai completions
...
closes #4505
2026-05-18 00:14:31 +02:00
266234047a
Remove openai-codex fast model variants, they do not work
2026-05-18 00:02:52 +02:00
b256ac7d77
Closes #4342
2026-05-17 23:55:25 +02:00
7f3c340dc6
Add [Unreleased] section for next cycle
2026-05-17 21:03:41 +02:00
a01cf7afae
Merge pull request #4603 from mattiacerutti/fix/openai-codex-model-list
...
fix(ai): update OpenAI Codex model list
2026-05-17 20:54:53 +02:00
6730c04a65
Merge pull request #4622 from mattiacerutti/fix/copilot-gpt-thinking-map-fix
...
fix(ai): map copilot gpt 5 minimal thinking to low
2026-05-17 20:53:42 +02:00
6d474f8c1a
fix(ai): cap context-sized default output budgets
...
closes #4614
2026-05-17 20:06:59 +02:00
485afc9c3f
fix(ai): map copilot gpt minimal thinking to low
2026-05-17 12:18:50 +02:00
1af823be9d
fix(ai): update OpenAI Codex model list
2026-05-17 03:12:19 +02:00
2c708492e3
Release v0.74.1
2026-05-17 01:35:45 +02:00
c5831df689
fix(ai): vendor proxy env resolution
...
closes #4513
2026-05-17 00:02:37 +02:00
22a9c484e7
fix(ai): respect model output token limits
...
closes #4539
2026-05-16 23:33:06 +02:00
7c5c3d6fd6
fix(ai): detect litellm context overflow errors
...
closes #4563
2026-05-16 23:29:22 +02:00
0412f62f9e
Merge pull request #4558 from earendil-works/fix/openai-completions-throw-on-missing-finish_reason
...
fix(ai): openai-completions - throw error on missing finish-reason
2026-05-16 22:54:42 +02:00
87881ca686
fix(ai): preserve OpenRouter cached token semantics
2026-05-16 11:55:16 +02:00
98ffad0437
fix(ai): openai-completions - throw error on missing finish-reason
...
- require \ before treating \ streams as successful
- add regression coverage for truncated streams without \
- closes #4345
2026-05-15 15:26:57 +02:00
a8af0b5e99
fix(ai): ignore generic GitHub tokens for Copilot auth
...
closes #4485
2026-05-15 01:26:31 +02:00
0ae909316a
fix(ai): honor retry-after for OpenAI Codex SSE retries
...
- honor `retry-after-ms` and `retry-after` for OpenAI Codex SSE retries
- add SSE retry coverage for millisecond, seconds, date, and fallback delays
2026-05-13 19:44:07 +02:00
40c05f5539
Merge pull request #4473 from apoorvumang/fix/inception-mercury-2-reasoning-off
...
fix(ai): mark inception/mercury-2 thinkingLevelMap.off as null
2026-05-13 18:48:49 +02:00
a5cca409d8
refactor(ai): use HTTP proxy agents for Bedrock
2026-05-13 16:12:48 +02:00
e2b69a0bb1
fix(ai): mark inception/mercury-2 thinkingLevelMap.off as null
...
Mercury 2 in instant mode (reasoning_effort: "none") disables tool calling.
The openai-completions provider hardcodes {reasoning:{effort:"none"}} when no
explicit reasoning level is passed and thinkingLevelMap.off isn't null
(openai-completions.ts:575), so every caller that doesn't opt in to a level
silently breaks Mercury 2's agentic use cases.
Setting thinkingLevelMap.off = null on the Mercury 2 catalog entry causes the
provider to omit the reasoning param entirely, letting Mercury 2's own default
take over. Low/medium/high pass through verbatim; OpenRouter normalizes them
to Mercury's vocabulary.
Prefix-matched on "inception/mercury-2" so future Mercury 2 variants on
OpenRouter inherit the fix.
2026-05-13 19:17:36 +05:30
f6b6b1f052
Merge pull request #4354 from haoqixu/fix-bun-ws-proxy
...
fix(ai): respect proxy envs in bun's websocket
2026-05-10 18:03:48 +02:00
99dc6fcec8
fix(ai): add session affinity and compat fixes for Fireworks provider caching
...
Fireworks prompt caching is enabled by default (automatic prefix matching),
but on serverless infrastructure, requests hit random replicas. Without
session affinity, the per-replica cache misses, negating cache hit rates
and the discounted cacheRead pricing.
Changes:
- Add sendSessionAffinityHeaders and supportsCacheControlOnTools
to AnthropicMessagesCompat interface
- Send x-session-affinity header for Fireworks (and Cloudflare AI
Gateway Anthropic) when sessionId is available and caching is enabled
- Omit cache_control on tool definitions for Fireworks (unsupported
per https://docs.fireworks.ai/tools-sdks/anthropic-compatibility )
- Default supportsEagerToolInputStreaming to false for Fireworks
(unsupported field)
- Default supportsLongCacheRetention to false for Fireworks
(cache_control.ttl not supported)
- Add compat settings to Fireworks models in generate-models.ts
- Update generated models with Fireworks compat settings
- Add integration tests for session affinity and tool compat
Refs: https://docs.fireworks.ai/guides/prompt-caching
Refs: https://docs.fireworks.ai/tools-sdks/anthropic-compatibility
2026-05-10 00:11:36 +02:00
8c2e3eddec
fix(ai): respect proxy envs in bun's websocket
...
fixes #4346
2026-05-10 03:43:07 +08:00
7adb8e7634
feat(ai): add Together AI provider
2026-05-08 16:44:18 +02:00
9751057be9
Merge pull request #3887 from cristinaponcela/feat/image-outputs
...
feat: image content
2026-05-08 15:57:06 +02:00
8fa1194c23
fix: no fallback for apiKey
2026-05-08 11:23:15 +02:00
0e4f845cf4
fix: lazy loading
2026-05-08 11:08:04 +02:00
783e96a144
fix(ai): disable OpenAI reasoning where supported
2026-05-07 22:52:54 +02:00
Armin Ronacher