Model comparison
Inkling-Small vs MiniMax-M2.7
Inkling-Small is the stronger model overall, scoring 46.5 to 37.7 on the Noometry Index.
Last verified . 22 shared benchmarks.
MiniMaxSummary
- They share 22 benchmarks with published results for both. Inkling-Small scores higher in 6 categories and MiniMax-M2.7 in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Inkling-Small leads 45.1 to 25.9.
- The biggest single-benchmark swing is CritPt: 8.3% for Inkling-Small and 0.6% for MiniMax-M2.7.
- MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.45 / $1.20 for Inkling-Small.
- Inkling-Small accepts more context: 524K tokens versus 205K.
Side by side
| Inkling-Small | MiniMax-M2.7 | |
|---|---|---|
| Provider | Thinking Machines Lab | MiniMax |
| Noometry Index | 46.5 | 37.7 |
| Released | 2026-07-15 | 2026-03-18 |
| Weights | Open | Open |
| Context window | 524K | 205K |
| Max output | 1.05M | 131K |
| Input $ / M tokens | $0.45 | $0.30 |
| Output $ / M tokens | $1.20 | $1.20 |
| Results tracked | 33 | 30 |
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-M2.7: 41.8 (#120)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| LMArena WebDev | 1409 | 1398 |
| SciCode | 48.7% | 47% |
| LMArena Coding | 1451 | 1454 |
| WeirdML | — | 37% |
| ALE-Bench | — | 599.25 |
Agentic & Tool Use Not comparable
Inkling-Small: —, MiniMax-M2.7: 25.1 (#111)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| Terminal-Bench | — | 45.1% |
| ExploitBench | — | 13.3% |
| GBAEval | — | 0% |
Reasoning Inkling-Small leads
Inkling-Small: 38.6 (#63), MiniMax-M2.7: 19.7 (#253)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| CritPt | 8.3% | 0.6% |
| LMArena Hard Prompts | 1423 | 1422 |
| Epoch Capabilities Index | 150.15 | 145.85 |
| ARC-AGI-2 | 40.1% | — |
| NYT Connections (extended) | — | 24.7% |
| ARC-AGI-1 | 84% | — |
| Chess Puzzles | 18% | — |
| Thematic Generalization | — | 39.3% |
| Mystery Game Puzzles | 6% | — |
Math Inkling-Small leads
Inkling-Small: 45.1 (#77), MiniMax-M2.7: 25.9 (#263)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| ProofBench | 6% | 3% |
| LMArena Math | 1459 | 1420 |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
| OTIS Mock AIME 2024-2025 | 90% | — |
Knowledge Inkling-Small leads
Inkling-Small: 48.2 (#77), MiniMax-M2.7: 37.7 (#152)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| LMArena Expert | 1442 | 1444 |
| GPQA Diamond | 88.5% | — |
| SimpleQA Verified | 19.1% | — |
| Vectara Hallucination Rate | — | 12.9% |
Multimodal Not comparable
Inkling-Small: 39.1 (#62), MiniMax-M2.7: —
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Inkling-Small leads
Inkling-Small: 51.7 (#104), MiniMax-M2.7: 50.3 (#123)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| LMArena Non-English | 1402 | 1382 |
| LMArena Chinese | 1465 | 1441 |
| LMArena French | 1436 | 1421 |
| LMArena German | 1405 | 1398 |
| LMArena Japanese | 1405 | 1262 |
| LMArena Korean | 1363 | 1313 |
| LMArena Russian | 1391 | 1383 |
| LMArena Spanish | 1428 | 1403 |
Instruction Following Too close to call
Inkling-Small: 73.8 (#114), MiniMax-M2.7: 74.1 (#103)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| LMArena Instruction Following | 1399 | 1405 |
Long Context Too close to call
Inkling-Small: 42.7 (#118), MiniMax-M2.7: 43.3 (#99)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| LMArena Longer Query | 1401 | 1419 |
Writing & Preference Too close to call
Inkling-Small: 59.6 (#107), MiniMax-M2.7: 58.9 (#112)
| Benchmark | Inkling-Small | MiniMax-M2.7 |
|---|---|---|
| LMArena Text | 1414 | 1405 |
| LMArena Creative Writing | 1331 | 1354 |
| LMArena Multi-Turn | 1418 | 1412 |
| EQ-Bench Creative Writing | 1491 | — |
Frequently asked questions
Is Inkling-Small better than MiniMax-M2.7?
Inkling-Small is the stronger model overall, scoring 46.5 to 37.7 on the Noometry Index.
Which is cheaper, Inkling-Small or MiniMax-M2.7?
MiniMax-M2.7 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-M2.7 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?
Inkling-Small does, with 524K tokens against 205K.
How many benchmarks do Inkling-Small and MiniMax-M2.7 share?
22 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and MiniMax-M2.7 has 30.