The provider-prefixed entries (deepseek/deepseek-chat, deepseek/deepseek-reasoner)
in the model cost map were missing supports_response_schema, supports_system_messages,
supports_native_streaming, supports_parallel_function_calling, and had stale
max_input_tokens / max_output_tokens values. This caused supports_response_schema()
to return False for DeepSeek models regardless of calling convention.
Changes:
- Sync deepseek/deepseek-chat and deepseek/deepseek-reasoner entries with
their canonical bare-name counterparts in both JSON files
- Add a bare-model-name fallback in _supports_factory so that when a
provider-prefixed entry is missing a capability field, the bare model
entry is consulted before returning False
- Fix pre-existing unused-import lint error (F401) in policy_resolve_endpoints.py
- Add 14 regression tests covering data consistency, API-level correctness,
and the new fallback logic
Add support for Alibaba Cloud's Qwen3-Max model with:
- 258K input tokens, 65K output tokens
- Tiered pricing based on context window usage (0-32K, 32K-128K, 128K-252K)
- Function calling and tool choice support
- Reasoning capabilities enabled
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Remove incorrect `:0` suffix from regional Bedrock model identifiers:
- us.anthropic.claude-opus-4-6-v1:0 → us.anthropic.claude-opus-4-6-v1
- au.anthropic.claude-opus-4-6-v1:0 (duplicate removed)
The `:0` suffix is invalid for Bedrock inference profile ARNs and causes
"The provided model identifier is invalid" errors when calling the model.
Fixes#20562
Notes: General support for Opus 4.6 was added in #20506 however
it omitted the AU (australian) specific instance profile used
in Bedrock. This change only adds the the au id. It is copied
from the US model settings which is consistent with past
additions of this regional model profile.
Unify follow-up fixes for Opus 4.6 pricing and routing metadata into
a single changeset.
Set long-context-capable Opus 4.6 entries to 1M input tokens where
>200K pricing is defined, align alias and dated capability metadata,
and add Bedrock Converse v1 IDs with and without :0 suffixes.
Keep regional endpoint pricing at a 10% premium over global entries
and mirror all cost-map changes in the backup file used for local
loading and offline fallback behavior.
Extend Opus 4.6 regression tests to verify metadata parity, Bedrock
regional pricing parity across :0 and non-:0 IDs, and converse model
registration in constants and runtime model sets.
* [Feat] add ElevenLabs `eleven_v3` and `eleven_multilingual_v2` to model cost map
Register ElevenLabs TTS models for cost tracking:
- elevenlabs/eleven_v3: most expressive model, 70+ languages, audio tags
- elevenlabs/eleven_multilingual_v2: default TTS model, 29 languages
Also update ElevenLabs docs with supported models table and eleven_v3 audio tags example.
* docs: remove model-agnostic tip from ElevenLabs docs
Add Claude Opus 4.6 entries for Anthropic, Bedrock Converse, and Vertex AI.
Align pricing and capability metadata with Anthropic docs, including
long-context rates, above-200k prompt-caching rates, prefill removal,
and tool-use system prompt token counts.
Register the Bedrock Converse model ID in constants and add targeted
tests to validate model map values and converse registration.
67 vercel_ai_gateway models were missing capability flags (supports_vision,
supports_function_calling, supports_tool_choice, supports_response_schema).
These capabilities were inferred from the corresponding direct provider entries
for the same models (e.g., vercel_ai_gateway/anthropic/claude-3.5-sonnet now has
the same capabilities as anthropic/claude-3.5-sonnet).
Models fixed include:
- Claude 3/3.5/3.7 (Anthropic)
- GPT-4/5 variants (OpenAI)
- Gemini 2.0/2.5 (Google)
- Grok 3/4 (xAI)
- Mistral/Mixtral variants
- Qwen models
- DeepSeek models
- And more
This ensures consistent capability reporting across providers for the same
underlying models.
Co-authored-by: krauckbot <krauckbot123@gmail.com>
Add moonshot/kimi-k2.5 model with:
- Input cost: $0.60/M tokens (6e-07)
- Output cost: $3.00/M tokens (3e-06)
- Cache read cost: $0.10/M tokens (1e-07)
- 256K context window
- Vision, function calling, tool choice, web search support
Reference: https://huggingface.co/moonshotai/Kimi-K2.5
Note: K2.5 thinking mode is controlled via API parameters, not a separate model ID.
Co-authored-by: krauckbot <krauckbot123@gmail.com>
Fixes#19788
- Add `supported_regions: ["global"]` to Qwen MaaS models in model_prices_and_context_window.json
- Update `get_supported_regions()` to read directly from `model_cost` dict
- Update `get_complete_vertex_url()` to use `get_vertex_region()` for global-only models
- Update `create_vertex_url()` to generate correct URL for global location (without region prefix)
- Add tests for Qwen global endpoint support
