Model comparison
Claude 3.5 Haiku vs MiniMax-M3
MiniMax-M3 is the stronger model overall, scoring 43.5 to 28.9 on the Noometry Index.
Last verified . 24 shared benchmarks.
MiniMaxSummary
- They share 24 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 1 category and MiniMax-M3 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 18.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 71.1% for MiniMax-M3.
- MiniMax-M3 has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Haiku | MiniMax-M3 | |
|---|---|---|
| Provider | Anthropic | MiniMax |
| Noometry Index | 28.9 | 43.5 |
| Released | 2024-10-22 | 2026-06-01 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 512K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 49 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M3 leads
Claude 3.5 Haiku: 32.6 (#258), MiniMax-M3: 41.5 (#111)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| SciCode | 27.4% | 47.1% |
| LMArena Coding | 1287 | 1470 |
| FrontierCode | — | 14.7% |
| Aider Polyglot | 28% | — |
| LMArena WebDev | — | 1482 |
| WeirdML | 30.7% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 640.02 |
Agentic & Tool Use Claude 3.5 Haiku leads
Claude 3.5 Haiku: 27.7 (#90), MiniMax-M3: 22.5 (#121)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| BALROG | 19.3% | — |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
Claude 3.5 Haiku: 17.1 (#286), MiniMax-M3: 29.3 (#83)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| CritPt | 0% | 3.7% |
| LMArena Hard Prompts | 1251 | 1447 |
| DTBench | 56.7% | 78.9% |
| Epoch Capabilities Index | 127.15 | 147 |
| SimpleBench | — | 45.8% |
| NYT Connections (extended) | — | 65.1% |
| Chess Puzzles | — | 14% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 8% |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 33.7% |
| Surface Evolver Bench | — | 55% |
| ForecastBench | — | 61.4 |
| LiveBench | 43.5% | — |
Math MiniMax-M3 leads
Claude 3.5 Haiku: 13.7 (#294), MiniMax-M3: 39.1 (#83)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 71.1% |
| LMArena Math | 1244 | 1432 |
| ProofBench | — | 18% |
| Omni-MATH | 22.4% | — |
| LiveBench Math | 35.5% | — |
| MATH Level 5 | 46.4% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge MiniMax-M3 leads
Claude 3.5 Haiku: 18.7 (#276), MiniMax-M3: 58.4 (#32)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | 38.1% | 90.9% |
| LMArena Expert | 1208 | 1462 |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| GPQA (HELM) | 36.3% | — |
| MMLU | 74.3% | — |
Multimodal MiniMax-M3 leads
Claude 3.5 Haiku: 26.3 (#113), MiniMax-M3: 39.5 (#45)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Vision | 1092 | 1254 |
| GeoBench | 34% | — |
| LMArena Document | — | 1435 |
Multilingual MiniMax-M3 leads
Claude 3.5 Haiku: 40.2 (#213), MiniMax-M3: 53.2 (#70)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1238 | 1420 |
| LMArena Chinese | 1229 | 1465 |
| LMArena French | 1262 | 1445 |
| LMArena German | 1237 | 1424 |
| LMArena Japanese | 1175 | 1381 |
| LMArena Korean | 1172 | 1372 |
| LMArena Russian | 1253 | 1430 |
| LMArena Spanish | 1261 | 1431 |
Instruction Following MiniMax-M3 leads
Claude 3.5 Haiku: 63.1 (#229), MiniMax-M3: 75.7 (#56)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1241 | 1433 |
| LiveBench Instruction Following | 61.9% | — |
| IFEval | 79.2% | — |
Long Context MiniMax-M3 leads
Claude 3.5 Haiku: 38.3 (#195), MiniMax-M3: 44.2 (#68)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1261 | 1445 |
Writing & Preference MiniMax-M3 leads
Claude 3.5 Haiku: 42.9 (#228), MiniMax-M3: 62.3 (#79)
| Benchmark | Claude 3.5 Haiku | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1255 | 1433 |
| LMArena Creative Writing | 1233 | 1405 |
| LMArena Multi-Turn | 1265 | 1441 |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | 1146 | — |
| WildBench | 76% | — |
| EQ-Bench 4 | — | 1150 |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than MiniMax-M3?
MiniMax-M3 is the stronger model overall, scoring 43.5 to 28.9 on the Noometry Index.
Is Claude 3.5 Haiku or MiniMax-M3 better for coding?
MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 32.6 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and MiniMax-M3 share?
24 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and MiniMax-M3 has 41.