Model comparison
Claude Sonnet 4 vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.5 to 40.6 on the Noometry Index.
Last verified . 28 shared benchmarks.
MiniMaxSummary
- They share 28 benchmarks with published results for both. Claude Sonnet 4 scores higher in 3 categories and MiniMax-M3 in 7 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 41.6.
- The biggest single-benchmark swing is GPQA Diamond: 79.2% for Claude Sonnet 4 and 90.9% for MiniMax-M3.
- MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- MiniMax-M3 accepts more context: 1M tokens versus 200K.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | MiniMax-M3 | |
|---|---|---|
| Provider | Anthropic | MiniMax |
| Noometry Index | 40.6 | 43.5 |
| Released | 2025-05-22 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | 200K | 1M |
| Max output | 64K | 512K |
| Input $ / M tokens | $3 | $0.30 |
| Output $ / M tokens | $15 | $1.20 |
| Results tracked | 58 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 44.0 (#70), MiniMax-M3: 41.5 (#111)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| SciCode | 40% | 47.1% |
| LMArena Coding | 1414 | 1470 |
| ALE-Bench | 655.35 | 640.02 |
| FrontierCode | — | 14.7% |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1482 |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.3 (#29), MiniMax-M3: 22.5 (#121)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| GBAEval | — | 0.9% |
| METR Time Horizons | 62% | — |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
Claude Sonnet 4: 23.0 (#169), MiniMax-M3: 29.3 (#83)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| SimpleBench | 45.5% | 45.8% |
| CritPt | 0.3% | 3.7% |
| LMArena Hard Prompts | 1372 | 1447 |
| DTBench | 77.1% | 78.9% |
| LMCA | 29% | 33.7% |
| Epoch Capabilities Index | 141.69 | 147 |
| ForecastBench | 60.2 | 61.4 |
| ARC-AGI-2 | 5.9% | — |
| Kagi LLM Benchmark | 73% | — |
| NYT Connections (extended) | — | 65.1% |
| ARC-AGI-1 | 40% | — |
| Chess Puzzles | — | 14% |
| EnigmaEval | 3.1% | — |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 41.5 (#71), MiniMax-M3: 39.1 (#83)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 71.1% |
| LMArena Math | 1375 | 1432 |
| ProofBench | — | 18% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge MiniMax-M3 leads
Claude Sonnet 4: 41.6 (#99), MiniMax-M3: 58.4 (#32)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 79.2% | 90.9% |
| LMArena Expert | 1371 | 1462 |
| Humanity's Last Exam | 7.8% | — |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal MiniMax-M3 leads
Claude Sonnet 4: 25.6 (#116), MiniMax-M3: 39.5 (#45)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1191 | 1254 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| LMArena Document | — | 1435 |
| MindCube | 44.8% | — |
Multilingual MiniMax-M3 leads
Claude Sonnet 4: 46.9 (#151), MiniMax-M3: 53.2 (#70)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1333 | 1420 |
| LMArena Chinese | 1351 | 1465 |
| LMArena French | 1364 | 1445 |
| LMArena German | 1331 | 1424 |
| LMArena Japanese | 1301 | 1381 |
| LMArena Korean | 1293 | 1372 |
| LMArena Russian | 1355 | 1430 |
| LMArena Spanish | 1356 | 1431 |
Instruction Following MiniMax-M3 leads
Claude Sonnet 4: 71.8 (#141), MiniMax-M3: 75.7 (#56)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1375 | 1433 |
| IFEval | 84% | — |
Long Context MiniMax-M3 leads
Claude Sonnet 4: 33.6 (#254), MiniMax-M3: 44.2 (#68)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1398 | 1445 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference MiniMax-M3 leads
Claude Sonnet 4: 57.3 (#127), MiniMax-M3: 62.3 (#79)
| Benchmark | Claude Sonnet 4 | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1351 | 1433 |
| LMArena Creative Writing | 1345 | 1405 |
| LMArena Multi-Turn | 1376 | 1441 |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is Claude Sonnet 4 better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.5 to 40.6 on the Noometry Index.
Which is cheaper, Claude Sonnet 4 or MiniMax-M3?
MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or MiniMax-M3 better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 44.0 versus 41.5 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 200K.
How many benchmarks do Claude Sonnet 4 and MiniMax-M3 share?
28 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and MiniMax-M3 has 41.