Model comparison
Inkling-Small vs Mixtral 8x22B
Inkling-Small is the stronger model overall, scoring 46.5 to 26.7 on the Noometry Index.
Last verified . 19 shared benchmarks.
Mistral AISummary
- They share 19 benchmarks with published results for both. Inkling-Small scores higher in 8 categories and Mixtral 8x22B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling-Small leads 48.3 to 15.1.
- The biggest single-benchmark swing is GPQA Diamond: 88.5% for Inkling-Small and 34.1% for Mixtral 8x22B.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Inkling-Small accepts more context: 524K tokens versus 64K.
Side by side
| Inkling-Small | Mixtral 8x22B | |
|---|---|---|
| Provider | Thinking Machines Lab | Mistral AI |
| Noometry Index | 46.5 | 26.7 |
| Released | 2026-07-15 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 524K | 64K |
| Max output | 1.05M | 64K |
| Input $ / M tokens | $0.45 | $2 |
| Output $ / M tokens | $1.20 | $6 |
| Results tracked | 33 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
Inkling-Small: 43.3 (#80), Mixtral 8x22B: 23.8 (#324)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| LMArena Coding | 1451 | 1166 |
| LMArena WebDev | 1408 | — |
| SciCode | 48.7% | — |
| WeirdML | — | 3.2% |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Not comparable
Inkling-Small: —, Mixtral 8x22B: 22.8 (#119)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| Cybench | — | 7.5% |
Reasoning Inkling-Small leads
Inkling-Small: 39.0 (#53), Mixtral 8x22B: 19.1 (#246)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1424 | 1150 |
| Epoch Capabilities Index | 150.13 | 122 |
| ARC-AGI-2 | 40.1% | — |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| Chess Puzzles | 18% | — |
| Mystery Game Puzzles | 6% | — |
| DTBench | — | 55.1% |
| ForecastBench | — | 56.3 |
Math Inkling-Small leads
Inkling-Small: 44.9 (#65), Mixtral 8x22B: 21.2 (#271)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1453 | 1184 |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
| OTIS Mock AIME 2024-2025 | 90% | — |
| ProofBench | 6% | — |
| Omni-MATH | — | 16.3% |
| MATH Level 5 | — | 24.2% |
Knowledge Inkling-Small leads
Inkling-Small: 48.3 (#71), Mixtral 8x22B: 15.1 (#288)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | 88.5% | 34.1% |
| LMArena Expert | 1442 | 1113 |
| SimpleQA Verified | 19.1% | — |
| MMLU-Pro | — | 46% |
| GPQA (HELM) | — | 33.4% |
| MMLU | — | 77.8% |
Multimodal Not comparable
Inkling-Small: 38.4 (#59), Mixtral 8x22B: —
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Inkling-Small leads
Inkling-Small: 51.7 (#102), Mixtral 8x22B: 32.9 (#250)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1400 | 1128 |
| LMArena Chinese | 1458 | 1116 |
| LMArena French | 1438 | 1165 |
| LMArena German | 1412 | 1141 |
| LMArena Japanese | 1399 | 1037 |
| LMArena Korean | 1355 | 1056 |
| LMArena Russian | 1392 | 1158 |
| LMArena Spanish | 1426 | 1150 |
Instruction Following Inkling-Small leads
Inkling-Small: 73.9 (#112), Mixtral 8x22B: 57.9 (#261)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1397 | 1147 |
| IFEval | — | 72.4% |
Long Context Inkling-Small leads
Inkling-Small: 42.7 (#113), Mixtral 8x22B: 34.6 (#242)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1401 | 1144 |
Writing & Preference Inkling-Small leads
Inkling-Small: 59.8 (#102), Mixtral 8x22B: 37.1 (#257)
| Benchmark | Inkling-Small | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1413 | 1162 |
| LMArena Creative Writing | 1330 | 1141 |
| LMArena Multi-Turn | 1420 | 1130 |
| EQ-Bench Creative Writing | 1491 | — |
| WildBench | — | 71.1% |
Frequently asked questions
Is Inkling-Small better than Mixtral 8x22B?
Inkling-Small is the stronger model overall, scoring 46.5 to 26.7 on the Noometry Index.
Which is cheaper, Inkling-Small or Mixtral 8x22B?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is Inkling-Small or Mixtral 8x22B better for coding?
Inkling-Small scores higher on coding benchmarks: 43.3 versus 23.8 in the Noometry coding category.
Which has the bigger context window?
Inkling-Small does, with 524K tokens against 64K.
How many benchmarks do Inkling-Small and Mixtral 8x22B share?
19 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Mixtral 8x22B has 34.