DeepSeek V3 vs MiniMax M2
Updated October 21, 2026. 7 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
DeepSeek V3 scores 70.2 and MiniMax M2 45.3 on the Vibe Coding Index, but no coding benchmark measured both, so the gap is directional, not a ranking.
Pick MiniMax M2 when output tokens cost 1.0x less, or when the context window is larger, 205K against 164K tokens.
- Shared results
- 7
- DeepSeek V3 only
- 6
- MiniMax M2 only
- 42
- Like-for-like categories
- 1 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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, four public coding benchmarks
No call
70.2 against 45.3, from different benchmarks with none in common.
Directional only
Intelligence
AA Intelligence Index
No call
Only MiniMax M2 has a AA Intelligence Index result, 18.6.
Directional only
Web design
Design Arena: websites
MiniMax M2
MiniMax M2 leads, 1144 against 1120.
Same board, both measured
Long documents
Listed context window
MiniMax M2
MiniMax M2 takes 205K tokens against 164K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
MiniMax M2
DeepSeek V3 $0.0008, MiniMax M2 $0.0008 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
- DeepSeek V3 70.2 from 1 benchmark, MiniMax M2 45.3 from 2 benchmarks.
- Widest category
- Web design, 6.1 points apart on average over 7 shared benchmarks.
- Widest single gap
- Design Arena: SVG: DeepSeek V3 976 against MiniMax M2 1098, 15.3 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V3 $1.0287 against MiniMax M2 $1.02 per million output tokens, 1.0x apart.
- Context window
- DeepSeek V3 takes 164K tokens, MiniMax M2 205K.
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 | DeepSeek V3 | Bars | MiniMax M2 | Ties |
|---|---|---|---|---|
| Web design | 0 | 7 | ||
| All shared boards | 0 | 7 |
CodingDirectional onlyFour of these boards feed the Vibe Coding Index10 boards, none shared
| Benchmark | DeepSeek V3 | MiniMax M2 | Leader |
|---|---|---|---|
| Aider polyglot | 70.2% | Not measured | Only DeepSeek V3 measured |
| SWE-bench Verified | Not measured | 61.0% | Only MiniMax M2 measured |
| SWE-bench Lite | 30.7% | Not measured | Only DeepSeek V3 measured |
| LMArena WebDev | Not measured | 1297 | Only MiniMax M2 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 82.6% | Only MiniMax M2 measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 25.8% | Only MiniMax M2 measured |
| Terminal-Bench | Not measured | 30.0% | Only MiniMax M2 measured |
| LMArena Text: coding | Not measured | 1371 | Only MiniMax M2 measured |
| LMArena Text: software and IT | Not measured | 1373 | Only MiniMax M2 measured |
| LMArena WebDev: HTML | Not measured | 1306 | Only MiniMax M2 measured |
| Neither model has a result on the other 28 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | DeepSeek V3 | MiniMax M2 | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 18.6 | Only MiniMax M2 measured |
| Neither model has a result on the other 1 board here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | DeepSeek V3 | MiniMax M2 | Leader |
|---|---|---|---|
| τ²-Bench Telecom | Not measured | 86.8% | Only MiniMax M2 measured |
| Vending-Bench 2 | Not measured | $161 | Only MiniMax M2 measured |
| Neither model has a result on the other 37 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | DeepSeek V3 | MiniMax M2 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 77.7% | Only MiniMax M2 measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 13.7% | Only MiniMax M2 measured |
| GPQA Diamond (Epoch AI run) | 71.2% | Not measured | Only DeepSeek V3 measured |
| LMArena Text: hard prompts | Not measured | 1357 | Only MiniMax M2 measured |
| LMArena Text: expert | Not measured | 1338 | Only MiniMax M2 measured |
| LMArena Text: hard prompts in English | Not measured | 1376 | Only MiniMax M2 measured |
| Neither model has a result on the other 15 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | DeepSeek V3 | MiniMax M2 | Leader |
|---|---|---|---|
| AA Math Index | Not measured | 78.3 | Only MiniMax M2 measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 78.3% | Only MiniMax M2 measured |
| OTIS Mock AIME | 48.9% | Not measured | Only DeepSeek V3 measured |
| LMArena Text: math | Not measured | 1351 | Only MiniMax M2 measured |
| LMArena Text: mathematical work | Not measured | 1368 | Only MiniMax M2 measured |
| Neither model has a result on the other 8 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | DeepSeek V3 | MiniMax M2 | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | Not measured | 82.0% | Only MiniMax M2 measured |
| LMArena Text: science | Not measured | 1348 | Only MiniMax M2 measured |
| LMArena Text: medicine and healthcare | Not measured | 1367 | Only MiniMax M2 measured |
| LMArena Text: legal and government | Not measured | 1349 | Only MiniMax M2 measured |
| LMArena Text: business and finance | Not measured | 1336 | Only MiniMax M2 measured |
| Neither model has a result on the other 12 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index07
| Benchmark | DeepSeek V3 | MiniMax M2 | Leader |
|---|---|---|---|
| Design Arena: websites | 1120 | 1144 | MiniMax M2 |
| Design Arena: UI components | 1093 | 1136 | MiniMax M2 |
| Design Arena: game dev | 1054 | 1119 | MiniMax M2 |
| Design Arena: data viz | 1089 | 1140 | MiniMax M2 |
| Design Arena: 3D | 1087 | 1098 | MiniMax M2 |
| Design Arena: all code categories | 1108 | 1135 | MiniMax M2 |
| Design Arena: SVG | 976 | 1098 | MiniMax M2 |
| Neither model has a result on the other 2 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index19 boards, none shared
| Benchmark | DeepSeek V3 | MiniMax M2 | Leader |
|---|---|---|---|
| IFBench | Not measured | 72.3% | Only MiniMax M2 measured |
| Long context reasoning (AA-LCR) | Not measured | 64.3% | Only MiniMax M2 measured |
| DTBench | 62.7% | Not measured | Only DeepSeek V3 measured |
| LMCA | 15.2% | Not measured | Only DeepSeek V3 measured |
| LMArena Text | Not measured | 1340 | Only MiniMax M2 measured |
| LMArena Text: instruction following | Not measured | 1329 | Only MiniMax M2 measured |
| LMArena Text: creative writing | Not measured | 1286 | Only MiniMax M2 measured |
| LMArena Text: English | Not measured | 1364 | Only MiniMax M2 measured |
| LMArena Text: non-English | Not measured | 1313 | Only MiniMax M2 measured |
| LMArena Text: Chinese | Not measured | 1366 | Only MiniMax M2 measured |
| LMArena Text: German | Not measured | 1357 | Only MiniMax M2 measured |
| LMArena Text: French | Not measured | 1336 | Only MiniMax M2 measured |
| LMArena Text: Spanish | Not measured | 1326 | Only MiniMax M2 measured |
| LMArena Text: Russian | Not measured | 1330 | Only MiniMax M2 measured |
| LMArena Text: Polish | Not measured | 1312 | Only MiniMax M2 measured |
| LMArena Text: multi-turn | Not measured | 1361 | Only MiniMax M2 measured |
| LMArena Text: longer queries | Not measured | 1331 | Only MiniMax M2 measured |
| LMArena Text: writing and language | Not measured | 1295 | Only MiniMax M2 measured |
| LMArena Text: entertainment and media | Not measured | 1298 | Only MiniMax M2 measured |
| Neither model has a result on the other 27 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V3 | MiniMax M2 | 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
MiniMax M2 costs less
Repository review
50K input + 3K output tokens
MiniMax M2 costs less
Agent loop
200K input + 10K output tokens
MiniMax M2 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 | DeepSeek V3 | MiniMax M2 |
|---|---|---|
| Maker | DeepSeek | MiniMax |
| Weights | Open weights | Open weights |
| Released | December 26, 2024 | October 23, 2025 |
| Context window | 164K | 205K |
| Input $/Mtok | $0.2574 | $0.255 |
| Output $/Mtok | $1.0287 | $1.02 |
| Coding index | 70.2 | 45.3 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | Not computable | $0.702 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V3 better than MiniMax M2 for coding?
DeepSeek V3 scores 70.2 and MiniMax M2 45.3 on the Vibe Coding Index, but no coding benchmark measured both, so the gap is directional, not a ranking. Pick MiniMax M2 when output tokens cost 1.0x less, or when the context window is larger, 205K against 164K tokens.
Which is cheaper, DeepSeek V3 or MiniMax M2?
DeepSeek V3 costs $0.2574 per million input tokens and $1.0287 per million output tokens; MiniMax M2 costs $0.255 and $1.02. Cost per solved task is known for MiniMax M2 only.
Which has the bigger context window, DeepSeek V3 or MiniMax M2?
MiniMax M2, with 205K tokens against 164K.
How many benchmarks is the DeepSeek V3 vs MiniMax M2 comparison based on?
7 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3's coding index rests on 1 independent benchmark and MiniMax M2's on 2. Where only one model has a result, the page says so instead of naming a winner.