W&B Inference (CoreWeave) Provider
Weights & Biases Inference (powered by CoreWeave) provides OpenAI-compatible inference for open-weight models — including DeepSeek V4 Flash, DeepSeek R1, Llama, and Qwen — at very competitive prices.
Available Models
DeepSeek V4 Flash (W&B Inference)
deepseek
wandb-deepseek-v4-flashStreaming
Tools
Reasoning
JSON Output
JSON Schema
W&B Inference (CoreWeave)
Context: 1M
Input
$0.01
/M tokens
Cached
—
/M tokens
Output
$0.01
/M tokens
TCS Cheap
tcs
tcs-cheapStreaming
Tools
Reasoning
JSON Output
JSON Schema
W&B Inference (CoreWeave)
Context: 1M
Input
$0.01
/M tokens
Cached
—
/M tokens
Output
$0.01
/M tokens
TCS Balanced
tcs
tcs-balancedStreaming
Tools
Reasoning
JSON Output
JSON Schema
W&B Inference (CoreWeave)
Context: 1M
Input
$0.01
/M tokens
Cached
—
/M tokens
Output
$0.01
/M tokens
TCS Reasoning
tcs
tcs-reasoningStreaming
Vision
Tools
Reasoning
JSON Output
W&B Inference (CoreWeave)
Context: 262.1k
Input
$0.95
/M tokens
Cached
$0.16
/M tokens
Output
$4
/M tokens
TCS Premium
tcs
tcs-premiumStreaming
Vision
Tools
Reasoning
JSON Output
W&B Inference (CoreWeave)
Context: 200k
Input
$1.4
/M tokens
Cached
$0.26
/M tokens
Output
$4.4
/M tokens
TCS Vision
tcs
tcs-visionStreaming
Vision
Tools
Reasoning
JSON Output
W&B Inference (CoreWeave)
Context: 262.1k
Input
$0.95
/M tokens
Cached
$0.16
/M tokens
Output
$4
/M tokens
Kimi K2.6 (W&B Inference)
moonshot
wandb-kimi-k2.6Streaming
Vision
Tools
Reasoning
JSON Output
W&B Inference (CoreWeave)
Context: 262.1k
Input
$0.95
/M tokens
Cached
$0.16
/M tokens
Output
$4
/M tokens
Z.AI GLM 5.1 (W&B Inference)
glm
wandb-glm-5.1Streaming
Vision
Tools
Reasoning
JSON Output
W&B Inference (CoreWeave)
Context: 200k
Input
$1.4
/M tokens
Cached
$0.26
/M tokens
Output
$4.4
/M tokens
Qwen3 235B A22B Thinking-2507 (W&B Inference)
qwen
wandb-qwen3-235b-thinkingStreaming
Tools
Reasoning
JSON Output
W&B Inference (CoreWeave)
Context: 262.1k
Input
$0.1
/M tokens
Cached
—
/M tokens
Output
$0.1
/M tokens
Llama 4 Scout 17Bx16E (W&B Inference)
llama
wandb-llama-4-scoutStreaming
Vision
Tools
JSON Output
W&B Inference (CoreWeave)
Context: 1.0M
Input
$0.17
/M tokens
Cached
—
/M tokens
Output
$0.66
/M tokens