Model comparison
Inkling-Small vs MiniMax-M2.5
Inkling-Small is the stronger model overall, scoring 46.5 to 38.3 on the Noometry Index.
Last verified . 23 shared benchmarks.
MiniMaxSummary
- They share 23 benchmarks with published results for both. Inkling-Small scores higher in 7 categories and MiniMax-M2.5 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling-Small leads 38.6 to 17.5.
- The biggest single-benchmark swing is ARC-AGI-2: 40.1% for Inkling-Small and 4.9% for MiniMax-M2.5.
- MiniMax-M2.5 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-M2.5 | |
|---|---|---|
| Provider | Thinking Machines Lab | MiniMax |
| Noometry Index | 46.5 | 38.3 |
| Released | 2026-07-15 | 2026-02-12 |
| 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 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Inkling-Small: 43.6 (#85), MiniMax-M2.5: 48.1 (#58)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| LMArena WebDev | 1409 | 1387 |
| LMArena Coding | 1451 | 1381 |
| SWE-bench Verified (bash only) | — | 75.8% |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 48.7% | — |
| ALE-Bench | — | 618.17 |
Agentic & Tool Use Not comparable
Inkling-Small: —, MiniMax-M2.5: 30.4 (#77)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Vending-Bench 2 | — | -23.16 |
Reasoning Inkling-Small leads
Inkling-Small: 38.6 (#63), MiniMax-M2.5: 17.5 (#292)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| ARC-AGI-2 | 40.1% | 4.9% |
| ARC-AGI-1 | 84% | 63.7% |
| LMArena Hard Prompts | 1423 | 1372 |
| Epoch Capabilities Index | 150.15 | 146.68 |
| Kagi LLM Benchmark | — | 55.2% |
| NYT Connections (extended) | — | 16.8% |
| CritPt | 8.3% | — |
| Chess Puzzles | 18% | — |
| Mystery Game Puzzles | 6% | — |
Math Inkling-Small leads
Inkling-Small: 45.1 (#77), MiniMax-M2.5: 26.9 (#253)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| ProofBench | 6% | 4% |
| LMArena Math | 1459 | 1378 |
| 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-M2.5: 39.2 (#135)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Expert | 1442 | 1379 |
| GPQA Diamond | 88.5% | — |
| SimpleQA Verified | 19.1% | — |
| Vectara Hallucination Rate | — | 9.1% |
Multimodal Not comparable
Inkling-Small: 39.1 (#62), MiniMax-M2.5: —
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Vision | 1235 | — |
Multilingual Inkling-Small leads
Inkling-Small: 51.7 (#104), MiniMax-M2.5: 47.1 (#152)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1402 | 1338 |
| LMArena Chinese | 1465 | 1393 |
| LMArena French | 1436 | 1362 |
| LMArena German | 1405 | 1362 |
| LMArena Japanese | 1405 | 1171 |
| LMArena Korean | 1363 | 1232 |
| LMArena Russian | 1391 | 1358 |
| LMArena Spanish | 1428 | 1354 |
Instruction Following Inkling-Small leads
Inkling-Small: 73.8 (#114), MiniMax-M2.5: 71.5 (#148)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1399 | 1353 |
Long Context Inkling-Small leads
Inkling-Small: 42.7 (#118), MiniMax-M2.5: 37.5 (#216)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1401 | 1366 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
Writing & Preference Inkling-Small leads
Inkling-Small: 59.6 (#107), MiniMax-M2.5: 53.9 (#153)
| Benchmark | Inkling-Small | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1414 | 1359 |
| LMArena Creative Writing | 1331 | 1331 |
| EQ-Bench Creative Writing | 1491 | 1361 |
| LMArena Multi-Turn | 1418 | 1364 |
Frequently asked questions
Is Inkling-Small better than MiniMax-M2.5?
Inkling-Small is the stronger model overall, scoring 46.5 to 38.3 on the Noometry Index.
Which is cheaper, Inkling-Small or MiniMax-M2.5?
MiniMax-M2.5 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-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 43.6 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-M2.5 share?
23 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and MiniMax-M2.5 has 33.