Compare, 2 models
Side by side.
Inkling Small and Mistral Large 4 on every board at least two of them share. Blue marks the best in each row.
Specs
List prices, context and published output speed. Blue marks the best in each row.
| Spec | Inkling SmallThinking Machines | Mistral Large 4Mistral |
|---|---|---|
| Lab | Thinking Machines | Mistral |
| Type | Language | Language |
| Released | Jul 30, 2026 | Oct 6, 2026 |
| Input price$ per 1M tokens | $0.30 | $1.36 |
| Output price$ per 1M tokens | $1.20 | $4.18 |
| Blended price3 input tokens per output | $0.52 | $2.07 |
| Contexttokens | 524.3K | 1.05M |
| Open weights | Yes | No |
| Output speedTok / s | 220 | n/a |
Benchmarks
Every board where at least two of these models have a published score. Each score links to its source.
| Board | Inkling SmallThinking Machines | Mistral Large 4Mistral |
|---|---|---|
| Epoch Capabilities IndexOverall, higher is better | 150 | 150 |
| LMArena textOverall, higher is better | 1406style control | 1432style control |
| FrontierMath Tiers 1-3Reasoning and knowledge, higher is better | 46.3%xhigh setting | 58.6%high setting |
| GPQA DiamondReasoning and knowledge, higher is better | 88.5%xhigh setting | 88.1%high setting |
n/a means the board has not published a score for that model.