Inkling
Thinking Machines
Key facts · verified Oct 30, 2026
Inkling is a Thinking Machines model. Its API price is $1 per million input tokens and $4.05 per million output tokens. Its context window is 256K tokens. It ranks #92 on LMArena's text leaderboard (Elo 1442, read 2 Nov 2026). Source: tinker-docs.thinkingmachines.ai.
Inkling is a model from Thinking Machines. Thinking Machines lists serverless inference pricing of $1 per million input tokens and $4.05 per million output tokens, with a 256K-token context window on its hosted service. The weights are available under the Apache 2.0 licence, so it can be self-hosted. On LMArena's text leaderboard it holds 1442 Elo (rank 92).
Pricing & specs checkedSource: vendor documentation · checked Oct 30, 2026LMArena: 1442 Elo (rank 92) · source · retrieved Nov 2, 2026
Full Specifications
$1 / $4.05 per million tokens | |
|---|---|
| Identity | |
| Developer | Thinking Machines |
| Licence | Apache 2.0 |
| Self-hostable | Yes |
| Cost | |
| Blended $/1M tokensinput × 0.75 + output × 0.25 | $1.762 / 1M tokens |
| Input price | $1.000 / 1M tokens |
| Output price | $4.050 / 1M tokens |
| Capacity | |
| Context window | 256K tokens |
| Measured quality | |
| LMArena Elo | 1442 (checked Nov 2026) |
| LMArena rank | Rank 92 |
| GPQA Diamond | 88 · xhigh effort |
| Elo per dollarLMArena Elo ÷ blended $/1M tokens | 818 |