Operator
$10 monthly allowance$9 / mo
$10 of Gateway credit, renewed each month.
Pay as you go after the credit.
- Works with Claude Code and Codex
- Automatic provider failover
- Personal usage meter
- Cancel any time
Inference for developers
Inference.net is an inference provider built for open weight models. Fast, stable, and affordable. Use our API to power your favorite apps and agents.
Or read the documentation| Model | Input | Cache read | Output | Model ID |
|---|---|---|---|---|
| Kimi K3 | $3.00 | $0.30 | $15.00 | |
| Kimi K3 Fast | $4.50 | $0.45 | $22.50 | |
| GLM-5.2 | $1.30 | $0.20 | $4.40 | |
| DeepSeek V4 Flash (0731) | $0.53 | $0.017 | $1.58 | |
| Claude Opus 5 Fast | $10.00 | $1.00 | $50.00 | |
| GPT-5.6 Sol Fast | $5.00 | $0.50 | $30.00 |
What is Inference.net?
Inference.net provides access to low-cost, high-performance AI models to power AI applications and coding agents.
- OpenAI-compatible Standard endpoints for any app or coding agent, including Claude Code and Codex
- One API key Every supported provider, without a separate key to create or rotate for each
- Automatic failover Each model routes across providers and moves to the next healthy route if one slows or fails
- No markup Provider pricing passed through with no per-request markup
- Zero Data Retention Prompts and outputs aren't retained after each request by default
490.4K tokens$4.47
More tokens, no limits.
Up to 10x more tokens than your Claude or Codex subscription. Never let a subscription limit break your focus and slow you down.
View plansKimi-K3-FastOpus 5
- SWE7274
- GPQA8183
- tok/s18438
- $/1M in4.55
Smart, fast, & affordable.
Fast inference on the top open-source models. Just as smart as Claude and Codex at a fraction of the cost. And did we mention much faster?
View integrations- >kimi-k3-fast[open]
- glm-5.2-fast[open]
- deepseek-v4-fast[open]
- opus-5[closed]
- gpt-5.6[closed]
- fable-5[closed]
All the best models.
One click to switch between the best open-source models and closed-source models like Claude and Codex. Find the right model for you.
Explore modelsYour data is your data.
Prompts and outputs are never used to train models. Data retention is kept to the minimum needed to operate the service.
- No training Your data never enters a training set.
- Scoped access Every key belongs to one project.
- Clear usage See exactly where tokens and spend go.
- Easy revocation Disable a credential in one click.
FAQ
Fast frontier models.
Kimi, GLM, DeepSeek, Fable, and GPT — routed through Inference.net. Open the full catalog for every model we serve.
Simple, predictable pricing
Studio
$60 monthly allowance$49 / mo
$60 of Gateway credit, renewed each month.
Pay as you go after the credit.
- Everything in Operator
- More monthly Gateway allowance
- Hosted and provider model access
- Cancel any time