MiniMax M2 vs Qwen3 235B A22B Instruct 2507
MiniMax M2 leads Qwen3 235B A22B Instruct 2507 on the Artificial Analysis Intelligence Index, 18.6 to 12.0, and Qwen3 235B A22B Instruct 2507 costs 1.3x less per output token ($0.775 vs $1.02 per million). Updated October 22, 2026; 17 benchmarks have a result for both.
MiniMax M2 or Qwen3 235B A22B Instruct 2507: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick MiniMax M2 for work measured by GPQA Diamond (Artificial Analysis run) or Humanity's Last Exam (Artificial Analysis run), where it leads.
- Shared results
- 17
- MiniMax M2 only
- 33
- Qwen3 235B A22B Instruct 2507 only
- 7
- Like-for-like categories
- 8 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
MiniMax M2
MiniMax M2 leads, 18.6 against 12.0.
Same board, both measured
Web design
Design Arena: websites
MiniMax M2
MiniMax M2 leads, 1143 against 1058.
Same board, both measured
Long documents
Listed context window
Qwen3 235B A22B Instruct 2507
Qwen3 235B A22B Instruct 2507 takes 262K tokens against 205K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 235B A22B Instruct 2507
MiniMax M2 $0.0008, Qwen3 235B A22B Instruct 2507 $0.0005 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only MiniMax M2 has the number, $0.702.
Directional only
At a glance
- Coding index
- MiniMax M2 not benchmarked, Qwen3 235B A22B Instruct 2507 not benchmarked.
- Widest category
- Agentic work, 33.6 points apart on average over 1 shared benchmark.
- Widest single gap
- τ²-Bench Telecom: MiniMax M2 86.8% against Qwen3 235B A22B Instruct 2507 53.2%, 33.6 points apart on a 0 to 100 scale.
- Output price
- MiniMax M2 $1.02 against Qwen3 235B A22B Instruct 2507 $0.775 per million output tokens, 1.3x apart.
- Context window
- MiniMax M2 takes 205K tokens, Qwen3 235B A22B Instruct 2507 262K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | MiniMax M2 | Bars | Qwen3 235B A22B Instruct 2507 | Ties |
|---|---|---|---|---|
| Coding | 2 | 0 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 1 | 0 | ||
| Reasoning | 0 | 2 | ||
| Math | 0 | 2 | ||
| Knowledge | 0 | 1 | ||
| Web design | 6 | 0 | ||
| General | 1 | 1 | ||
| All shared boards | 11 | 6 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index20
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| SWE-bench Verified | 61.0% | Not measured | Only MiniMax M2 measured |
| LMArena WebDev | 1297 | Not measured | Only MiniMax M2 measured |
| LiveCodeBench (Artificial Analysis run) | 82.6% | 78.8% | MiniMax M2 |
| SciCode | Not measured | 41.4% | Only Qwen3 235B A22B Instruct 2507 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 25.8% | 15.2% | MiniMax M2 |
| Terminal-Bench | 30.0% | Not measured | Only MiniMax M2 measured |
| WeirdML | Not measured | 38.7% | Only Qwen3 235B A22B Instruct 2507 measured |
| LMArena Text: coding | 1371 | Not measured | Only MiniMax M2 measured |
| LMArena Text: software and IT | 1373 | Not measured | Only MiniMax M2 measured |
| LMArena WebDev: HTML | 1306 | Not measured | Only MiniMax M2 measured |
| Neither model has a result on the other 33 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| AA Intelligence Index | 18.6 | 12.0 | MiniMax M2 |
| Neither model has a result on the other 1 board here | |||
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 86.8% | 53.2% | MiniMax M2 |
| Vending-Bench 2 | $161 | Not measured | Only MiniMax M2 measured |
| Berkeley Function Calling Leaderboard (BFCL) | Not measured | 52.1% | Only Qwen3 235B A22B Instruct 2507 measured |
| Neither model has a result on the other 39 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 77.7% | 79.0% | Qwen3 235B A22B Instruct 2507 |
| Humanity's Last Exam (Artificial Analysis run) | 13.7% | 15.9% | Qwen3 235B A22B Instruct 2507 |
| ARC-AGI | Not measured | 11.0% | Only Qwen3 235B A22B Instruct 2507 measured |
| ARC-AGI-2 | Not measured | 1.3% | Only Qwen3 235B A22B Instruct 2507 measured |
| LMArena Text: hard prompts | 1357 | Not measured | Only MiniMax M2 measured |
| LMArena Text: expert | 1338 | Not measured | Only MiniMax M2 measured |
| LMArena Text: hard prompts in English | 1376 | Not measured | Only MiniMax M2 measured |
| Kagi LLM Benchmark | 57.8% | Not measured | Only MiniMax M2 measured |
| Neither model has a result on the other 15 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| AA Math Index | 78.3 | 91.0 | Qwen3 235B A22B Instruct 2507 |
| AIME 2025 (Artificial Analysis run) | 78.3% | 91.0% | Qwen3 235B A22B Instruct 2507 |
| LMArena Text: math | 1351 | Not measured | Only MiniMax M2 measured |
| LMArena Text: mathematical work | 1368 | Not measured | Only MiniMax M2 measured |
| Neither model has a result on the other 10 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 82.0% | 84.3% | Qwen3 235B A22B Instruct 2507 |
| LMArena Text: science | 1348 | Not measured | Only MiniMax M2 measured |
| LMArena Text: medicine and healthcare | 1367 | Not measured | Only MiniMax M2 measured |
| LMArena Text: legal and government | 1349 | Not measured | Only MiniMax M2 measured |
| LMArena Text: business and finance | 1336 | Not measured | Only MiniMax M2 measured |
| Neither model has a result on the other 15 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| Design Arena: websites | 1143 | 1058 | MiniMax M2 |
| Design Arena: UI components | 1134 | 963 | MiniMax M2 |
| Design Arena: game dev | 1118 | 951 | MiniMax M2 |
| Design Arena: data viz | 1140 | 1070 | MiniMax M2 |
| Design Arena: 3D | 1098 | 994 | MiniMax M2 |
| Design Arena: all code categories | 1135 | 1038 | MiniMax M2 |
| Design Arena: SVG | 1100 | Not measured | Only MiniMax M2 measured |
| Neither model has a result on the other 2 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| IFBench | 72.3% | 51.2% | MiniMax M2 |
| Long context reasoning (AA-LCR) | 64.3% | 72.0% | Qwen3 235B A22B Instruct 2507 |
| DTBench | Not measured | 78.4% | Only Qwen3 235B A22B Instruct 2507 measured |
| LMCA | Not measured | 23.5% | Only Qwen3 235B A22B Instruct 2507 measured |
| LMArena Text | 1340 | Not measured | Only MiniMax M2 measured |
| LMArena Text: instruction following | 1329 | Not measured | Only MiniMax M2 measured |
| LMArena Text: creative writing | 1286 | Not measured | Only MiniMax M2 measured |
| LMArena Text: English | 1364 | Not measured | Only MiniMax M2 measured |
| LMArena Text: non-English | 1313 | Not measured | Only MiniMax M2 measured |
| LMArena Text: Chinese | 1366 | Not measured | Only MiniMax M2 measured |
| LMArena Text: German | 1357 | Not measured | Only MiniMax M2 measured |
| LMArena Text: French | 1336 | Not measured | Only MiniMax M2 measured |
| LMArena Text: Spanish | 1326 | Not measured | Only MiniMax M2 measured |
| LMArena Text: Russian | 1330 | Not measured | Only MiniMax M2 measured |
| LMArena Text: Polish | 1312 | Not measured | Only MiniMax M2 measured |
| LMArena Text: multi-turn | 1361 | Not measured | Only MiniMax M2 measured |
| LMArena Text: longer queries | 1331 | Not measured | Only MiniMax M2 measured |
| LMArena Text: writing and language | 1295 | Not measured | Only MiniMax M2 measured |
| LMArena Text: entertainment and media | 1298 | Not measured | Only MiniMax M2 measured |
| Neither model has a result on the other 29 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | MiniMax M2 | Qwen3 235B A22B Instruct 2507 | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Qwen3 235B A22B Instruct 2507 costs less
Repository review
50K input + 3K output tokens
Qwen3 235B A22B Instruct 2507 costs less
Agent loop
200K input + 10K output tokens
Qwen3 235B A22B Instruct 2507 costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | MiniMax M2 | Qwen3 235B A22B Instruct 2507 |
|---|---|---|
| Maker | MiniMax | Qwen |
| Weights | Open weights | Open weights |
| Released | October 23, 2025 | July 21, 2025 |
| Context window | 205K | 262K |
| Input $/Mtok | $0.255 | $0.145 |
| Output $/Mtok | $1.02 | $0.775 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 1 | 0 |
| Cost per solved task | $0.702 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is MiniMax M2 better than Qwen3 235B A22B Instruct 2507 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick MiniMax M2 for work measured by GPQA Diamond (Artificial Analysis run) or Humanity's Last Exam (Artificial Analysis run), where it leads.
Which is cheaper, MiniMax M2 or Qwen3 235B A22B Instruct 2507?
MiniMax M2 costs $0.255 per million input tokens and $1.02 per million output tokens; Qwen3 235B A22B Instruct 2507 costs $0.145 and $0.775. Cost per solved task is known for MiniMax M2 only.
Which has the bigger context window, MiniMax M2 or Qwen3 235B A22B Instruct 2507?
Qwen3 235B A22B Instruct 2507, with 262K tokens against 205K.
How many benchmarks is the MiniMax M2 vs Qwen3 235B A22B Instruct 2507 comparison based on?
17 benchmarks have a result for both models as of October 22, 2026. MiniMax M2's coding index rests on 1 independent benchmark and Qwen3 235B A22B Instruct 2507's on 0. Where only one model has a result, the page says so instead of naming a winner.