Model comparison
GPT-3.5-turbo vs MiniMax-M2
MiniMax-M2 is the stronger model overall, scoring 37.4 to 23.2 on the Noometry Index.
Last verified . 15 shared benchmarks.
MiniMaxSummary
- They share 15 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and MiniMax-M2 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where MiniMax-M2 leads 37.3 to 6.3.
- MiniMax-M2 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.50 / $1.50 for GPT-3.5-turbo.
- MiniMax-M2 accepts more context: 205K tokens versus 16K.
- MiniMax-M2 has downloadable open weights; the other is API-only.
Side by side
| GPT-3.5-turbo | MiniMax-M2 | |
|---|---|---|
| Provider | OpenAI | MiniMax |
| Noometry Index | 23.2 | 37.4 |
| Released | 2023-03-01 | 2025-10-27 |
| Weights | Proprietary | Open |
| Context window | 16K | 205K |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.50 | $0.30 |
| Output $ / M tokens | $1.50 | $1.20 |
| Results tracked | 44 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2 leads
GPT-3.5-turbo: 23.9 (#331), MiniMax-M2: 39.3 (#159)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| LMArena Coding | 1136 | 1370 |
| SWE-bench Verified (bash only) | — | 61% |
| LMArena WebDev | — | 1297 |
| WeirdML | 3.5% | — |
| BigCodeBench Instruct | 39.1% | — |
| BigCodeBench Complete | 50.6% | — |
| HumanEval+ | 70.7% | — |
| MBPP+ | 69.7% | — |
Agentic & Tool Use Not comparable
GPT-3.5-turbo: —, MiniMax-M2: 25.1 (#109)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| Terminal-Bench | — | 30% |
| METR Time Horizons | 21.5% | — |
| Vending-Bench 2 | — | 160.6 |
Reasoning MiniMax-M2 leads
GPT-3.5-turbo: 13.8 (#332), MiniMax-M2: 19.4 (#258)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| LMArena Hard Prompts | 1108 | 1357 |
| Kagi LLM Benchmark | — | 57.8% |
| NYT Connections (extended) | — | 14.8% |
| Chess Puzzles | 0% | — |
| Mystery Game Puzzles | 3% | — |
| DTBench | 48.5% | — |
| LMCA | 9.7% | — |
| Adversarial NLI | 58.1% | — |
| BIG-Bench Hard | 61.6% | — |
| CommonsenseQA 2.0 | 57% | — |
| Epoch Capabilities Index | 118.55 | — |
| ForecastBench | 50.4 | — |
| WinoGrande | 81.6% | — |
Math MiniMax-M2 leads
GPT-3.5-turbo: 6.3 (#327), MiniMax-M2: 37.3 (#160)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| LMArena Math | 1142 | 1352 |
| FrontierMath (Tiers 1-3) | 0% | — |
| OTIS Mock AIME 2024-2025 | 2.2% | — |
| MATH Level 5 | 15.9% | — |
| GSM8K | 57.8% | — |
Knowledge MiniMax-M2 leads
GPT-3.5-turbo: 10.0 (#303), MiniMax-M2: 37.0 (#163)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| LMArena Expert | 1070 | 1337 |
| GPQA Diamond | 28% | — |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| MMLU | 71.4% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multilingual MiniMax-M2 leads
GPT-3.5-turbo: 31.5 (#258), MiniMax-M2: 45.3 (#171)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| LMArena Non-English | 1108 | 1313 |
| LMArena Chinese | 1075 | 1366 |
| LMArena French | 1118 | 1335 |
| LMArena German | 1090 | 1355 |
| LMArena Russian | 1123 | 1331 |
| LMArena Spanish | 1121 | 1326 |
| LMArena Japanese | 1043 | — |
| LMArena Korean | 1019 | — |
Instruction Following MiniMax-M2 leads
GPT-3.5-turbo: 57.9 (#262), MiniMax-M2: 70.2 (#166)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| LMArena Instruction Following | 1119 | 1328 |
Long Context MiniMax-M2 leads
GPT-3.5-turbo: 34.0 (#254), MiniMax-M2: 40.5 (#153)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| LMArena Longer Query | 1121 | 1331 |
Writing & Preference MiniMax-M2 leads
GPT-3.5-turbo: 25.3 (#305), MiniMax-M2: 53.0 (#162)
| Benchmark | GPT-3.5-turbo | MiniMax-M2 |
|---|---|---|
| LMArena Text | 1125 | 1340 |
| LMArena Creative Writing | 1092 | 1286 |
| LMArena Multi-Turn | 1117 | 1361 |
| EQ-Bench Creative Writing | 451 | — |
Frequently asked questions
Is GPT-3.5-turbo better than MiniMax-M2?
MiniMax-M2 is the stronger model overall, scoring 37.4 to 23.2 on the Noometry Index.
Which is cheaper, GPT-3.5-turbo or MiniMax-M2?
MiniMax-M2 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-3.5-turbo lists at $0.50 and $1.50.
Is GPT-3.5-turbo or MiniMax-M2 better for coding?
MiniMax-M2 scores higher on coding benchmarks: 39.3 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2 does, with 205K tokens against 16K.
How many benchmarks do GPT-3.5-turbo and MiniMax-M2 share?
15 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and MiniMax-M2 has 21.