# Thinking Machines models

[Get API key](https://vercel.com/d?to=%2F%5Bteam%5D%2F~%2Fai-gateway%3FshowCreateKeyModal%26utm_source%3Dgateway-providers-page%26utm_campaign%3Dai-gateway-providers&title=Get%20API%20key) [Read the docs](https://vercel.com/docs/ai-gateway)

Tinker is Thinking Machines Lab’s flexible platform for customizing open-weight models at any scale. Build specialized models with supervised fine-tuning or reinforcement learning while Tinker manages the distributed infrastructure, scheduling, and reliability. You stay focused on your data, algorithms, and ideas.

| Model | Context | Latency | Throughput | Input | Output | Cache | Web Search | Capabilities | ZDR | No Training | Regional Inference | Free Tier | Released |  |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |

| ![thinkingmachines logo](/vc-ap-vercel-marketing/_next/image?url=https%3A%2F%2F7nyt0uhk7sse4zvn.public.blob.vercel-storage.com%2Fdocs-assets%2Fstatic%2Fdocs%2Fai-gateway%2Flogos%2Fthinkingmachines.png%3Fv%3D1787369193129&w=48&q=75) [thinkingmachines/inkling-small](/ai-gateway/models/inkling-small) | 64K | 1.7s | 284tps | $0.30/M | $1.20/M | Read:$0.06/M Write:— | — | +1 |  |  |  |  | 07/30/2026 |  |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| ![thinkingmachines logo](/vc-ap-vercel-marketing/_next/image?url=https%3A%2F%2F7nyt0uhk7sse4zvn.public.blob.vercel-storage.com%2Fdocs-assets%2Fstatic%2Fdocs%2Fai-gateway%2Flogos%2Fthinkingmachines.png%3Fv%3D1787369193129&w=48&q=75) [thinkingmachines/inkling](/ai-gateway/models/inkling) | 256K | 2.9s | 116tps | $1/M | $4.05/M | Read:$0.17/M Write:— | — | +1 |  |  |  |  | 07/15/2026 |  |