Azure Foundry deploys gpt-5.4 and gpt-5.5 with a 1,050,000 context
window, but the Azure provider cloned OpenAI-direct's 272k API limit.
Override the context window in the Azure derivation.
Also fix gpt-5-pro maxTokens, which upstream metadata set to 272000 (a
duplicate of the input sub-limit) instead of the actual 128000 max
output; corrected at the source so OpenAI-direct and Azure both match.
closes#5559
Removed together.ai MiniMaxAI/MiniMax-M2.5 due to it not being supported for serverless inference.
Error: 400 Unable to access non-serverless model MiniMaxAI/MiniMax-M2.5. Please visit
https://api.together.ai/models/MiniMaxAI/MiniMax-M2.5 to create and start a new dedicated endpoint for the model.
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.
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
Built-in `xiaomi` provider now targets the API billing endpoint (https://api.xiaomimimo.com/anthropic) — a single stable URL for keys issued at platform.xiaomimimo.com. The Token Plan endpoints are exposed as three sibling providers, each with its own env var:
- xiaomi-token-plan-cn: XIAOMI_TOKEN_PLAN_CN_API_KEY
- xiaomi-token-plan-ams: XIAOMI_TOKEN_PLAN_AMS_API_KEY
- xiaomi-token-plan-sgp: XIAOMI_TOKEN_PLAN_SGP_API_KEY
BREAKING CHANGE: users who previously set XIAOMI_API_KEY against the Token Plan AMS endpoint must move to xiaomi-token-plan-ams and set XIAOMI_TOKEN_PLAN_AMS_API_KEY. This also resolves the 401 reported by on #4005, where a platform.xiaomimimo.com key fails against the Token Plan endpoint.
closes#4082
