Gateway
Weights & Biases
Weights & Biases is a Gateway provider with 8 models across 8 offerings.
Access
- API base URL
- https://api.inference.wandb.ai/v1
- SDK
- @ai-sdk/openai-compatible
- Environment variables
- WANDB_API_KEY
Models
8 of 8 model versions
| Model | Family | Provider model ID | Protocol | Streaming | Organization | Context | Max output | Input | Output | Reasoning | Tool call | Structured | Temperature | Weights | Price | Released | Updated | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Kimi K3 moonshotai/kimi-k3 | MAMoonshot AI | Kimi | moonshotai/Kimi-K3 | Openai chat | Yes | 1M | 131K | Yes | Yes | Yes | -No | Open | input $3/M · output $15/M cache read $0.3/M | 2026-07-16 | 2026-07-30 | |||
| Kimi K2.7 Code moonshotai/kimi-k2.7-code | MAMoonshot AI | Kimi | moonshotai/Kimi-K2.7-Code | Openai chat | Yes | 262K | 262K | Yes | Yes | Yes | -No | Open | input $0.71/M · output $3.5/M cache read $0.15/M | 2026-06-12 | 2026-08-06 | |||
| Granite 4.1 8B ibm-granite/granite-4.1-8b | IGIbm Granite | Granite | ibm-granite/granite-4.1-8b | Openai chat | Yes | 131K | 131K | -No | Yes | Yes | Yes | Open | input $0.05/M · output $0.1/M cache read $0.05/M | 2026-04-30 | 2026-08-06 | |||
| Kimi K2.6 moonshotai/kimi-k2.6 | MAMoonshot AI | Kimi | moonshotai/Kimi-K2.6 | Openai chat | Yes | 262K | 262K | Yes | Yes | Yes | Yes | Open | input $0.65/M · output $3.41/M cache read $0.15/M | 2026-04-21 | 2026-08-06 | |||
| Gemma 4 31B IT google/gemma-4-31b-it | GOGoogle | Gemma | google/gemma-4-31B-it | Openai chat | Yes | 262K | 33K | Yes | Yes | Yes | Yes | Open | input $0.1/M · output $0.34/M cache read $0.1/M | 2026-04-02 | 2026-08-06 | |||
| Kimi K2.5 moonshotai/kimi-k2.5 | MAMoonshot AI | Kimi | moonshotai/Kimi-K2.5 | Openai chat | Yes | 262K | 262K | Yes | Yes | Yes | -No | Open | input $0.6/M · output $3/M cache read $0.1/M | 2026-01 | 2026-08-06 | |||
| gpt-oss-120b openai/gpt-oss-120b | OPOpenAI | GPT | openai/gpt-oss-120b | Openai chat | Yes | 131K | 131K | Yes | Yes | Yes | Yes | Open | input $0.03/M · output $0.17/M cache read $0.03/M | 2025-08-05 | 2026-08-06 | |||
| gpt-oss-20b openai/gpt-oss-20b | OPOpenAI | GPT | openai/gpt-oss-20b | Openai chat | Yes | 131K | 131K | Yes | Yes | Yes | Yes | Open | input $0.03/M · output $0.13/M cache read $0.03/M | 2025-08-05 | 2026-08-06 |