W&B Inference

Ongoing verified 2026-08-14 · 0d ago no cardcommercial OKOpenAI-compatible

Open-weight frontier models (DeepSeek, Llama, Qwen, GLM, GPT-OSS) behind one unified API with Weave tracing

What's free

$100/month of Serverless Inference credits on the Free plan (default spending cap; offer for a limited time)

Rate limits

No RPM/TPM published; default cap of $100/month on the Free tier; concurrency limits per project/user

The catch

Serverless Inference credits come with Free, Pro and Academic plans for a limited time; when credits run out, Free accounts must activate pay-as-you-go on the Billing tab or upgrade. OpenAI-compatible endpoint at api.inference.wandb.ai/v1 with any W&B API key.

TypeOngoing free tier
Free typerecurring-credit
Expiresno expiry
Modalitiestext, vision
OpenAI base URLhttps://api.inference.wandb.ai/v1

Free models · sample

deepseek-ai/DeepSeek-V4-Flashmeta-llama/Llama-3.3-70B-Instructmeta-llama/Llama-3.1-8B-InstructMiniMaxAI/MiniMax-M3openai/gpt-oss-120bopenai/gpt-oss-20bzai-org/GLM-5.2google/gemma-4-31B-itnvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B

A sample of models reachable on the free tier — the live catalog changes. Pull the current set with GET https://api.inference.wandb.ai/v1/models.

Quickstart — chat completions

from openai import OpenAI

client = OpenAI(base_url="https://api.inference.wandb.ai/v1", api_key="<YOUR_FREE_API_KEY>")
resp = client.chat.completions.create(
    model="deepseek-ai/DeepSeek-V4-Flash",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)

…or with curl:

curl https://api.inference.wandb.ai/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-ai/DeepSeek-V4-Flash","messages":[{"role":"user","content":"Hello!"}]}'

Appears in

Change history

How this free tier has changed since we started tracking it (2026-08-14) — generated from the git history of providers.json.

← All providers