Model comparison
Inkling-Small vs MiniMax-M3
Inkling-Small is the stronger model overall, scoring 46.5 to 43.8 on the Noometry Index.
Last verified . 27 shared benchmarks.
MiniMaxSummary
- They share 27 benchmarks with published results for both. Inkling-Small scores higher in 3 categories and MiniMax-M3 in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 48.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 90% for Inkling-Small and 71.1% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.45 / $1.20 for Inkling-Small.
- MiniMax-M3 accepts more context: 1M tokens versus 524K.
Side by side
| Inkling-Small | MiniMax-M3 | |
|---|---|---|
| Provider | Thinking Machines Lab | MiniMax |
| Noometry Index | 46.5 | 43.8 |
| Released | 2026-07-15 | 2026-06-01 |
| Weights | Open | Open |
| Context window | 524K | 1M |
| Max output | 1.05M | 512K |
| Input $ / M tokens | $0.45 | $0.30 |
| Output $ / M tokens | $1.20 | $1.20 |
| Results tracked | 33 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Inkling-Small leads
Inkling-Small: 43.6 (#85), MiniMax-M3: 41.8 (#118)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| LMArena WebDev | 1409 | 1482 |
| SciCode | 48.7% | 47.1% |
| LMArena Coding | 1451 | 1469 |
| FrontierCode | — | 14.7% |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use Not comparable
Inkling-Small: —, MiniMax-M3: 22.6 (#130)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning Inkling-Small leads
Inkling-Small: 38.6 (#63), MiniMax-M3: 30.1 (#87)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| CritPt | 8.3% | 3.7% |
| Chess Puzzles | 18% | 14% |
| LMArena Hard Prompts | 1423 | 1447 |
| Mystery Game Puzzles | 6% | 8% |
| Epoch Capabilities Index | 150.15 | 146.95 |
| ARC-AGI-2 | 40.1% | — |
| SimpleBench | — | 45.8% |
| NYT Connections (extended) | — | 65.1% |
| ARC-AGI-1 | 84% | — |
| DTBench | — | 78.9% |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
Math Inkling-Small leads
Inkling-Small: 45.1 (#77), MiniMax-M3: 40.0 (#95)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 90% | 71.1% |
| ProofBench | 6% | 18% |
| LMArena Math | 1459 | 1429 |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
Knowledge MiniMax-M3 leads
Inkling-Small: 48.2 (#77), MiniMax-M3: 58.4 (#35)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 88.5% | 90.9% |
| LMArena Expert | 1442 | 1461 |
| SimpleQA Verified | 19.1% | — |
Multimodal MiniMax-M3 leads
Inkling-Small: 39.1 (#62), MiniMax-M3: 40.2 (#51)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1235 | 1253 |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
Inkling-Small: 51.7 (#104), MiniMax-M3: 53.0 (#75)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1402 | 1420 |
| LMArena Chinese | 1465 | 1463 |
| LMArena French | 1436 | 1447 |
| LMArena German | 1405 | 1426 |
| LMArena Japanese | 1405 | 1381 |
| LMArena Korean | 1363 | 1372 |
| LMArena Russian | 1391 | 1428 |
| LMArena Spanish | 1428 | 1432 |
Instruction Following MiniMax-M3 leads
Inkling-Small: 73.8 (#114), MiniMax-M3: 75.5 (#62)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1399 | 1433 |
Long Context MiniMax-M3 leads
Inkling-Small: 42.7 (#118), MiniMax-M3: 44.2 (#72)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1401 | 1445 |
Writing & Preference MiniMax-M3 leads
Inkling-Small: 59.6 (#107), MiniMax-M3: 62.1 (#83)
| Benchmark | Inkling-Small | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1414 | 1433 |
| LMArena Creative Writing | 1331 | 1404 |
| LMArena Multi-Turn | 1418 | 1442 |
| EQ-Bench Creative Writing | 1491 | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Inkling-Small better than MiniMax-M3?
Inkling-Small is the stronger model overall, scoring 46.5 to 43.8 on the Noometry Index.
Which is cheaper, Inkling-Small or MiniMax-M3?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Inkling-Small lists at $0.45 and $1.20.
Is Inkling-Small or MiniMax-M3 better for coding?
Inkling-Small scores higher on coding benchmarks: 43.6 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 524K.
How many benchmarks do Inkling-Small and MiniMax-M3 share?
27 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and MiniMax-M3 has 41.