Gemma 4 26B A4B vs MiniMax M3
MiniMax M3 leads Gemma 4 26B A4B on the Artificial Analysis Intelligence Index, 29.2 to 16.7, and Gemma 4 26B A4B costs 2.0x less per output token ($0.60 vs $1.20 per million). Updated October 22, 2026; 52 benchmarks have a result for both.
Gemma 4 26B A4B or MiniMax M3: which to pick
Only MiniMax M3 has a coding score, 39.2 on the Vibe Coding Index; Gemma 4 26B A4B has none, so the two cannot be ranked on coding.
Pick Gemma 4 26B A4B for output tokens 2.0x cheaper ($0.60 vs $1.20 per million) and work measured by ALE-Bench or LMArena Vision: homework, where it leads.
- Shared results
- 52
- Gemma 4 26B A4B only
- 3
- MiniMax M3 only
- 64
- Like-for-like categories
- 7 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
Only MiniMax M3 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
MiniMax M3
MiniMax M3 leads, 29.2 against 16.7.
Same board, both measured
Agentic work
AA Agentic Index
MiniMax M3
MiniMax M3 leads, 29.5 against 2.9.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only MiniMax M3 has a GPQA Diamond result, 92.7%.
Directional only
Knowledge
MMLU Pro
No call
Only MiniMax M3 has a MMLU Pro result, 84.2%.
Directional only
Web design
Design Arena: websites
No call
Only MiniMax M3 has a Design Arena: websites result, 1257.
Directional only
Long documents
Listed context window
MiniMax M3
MiniMax M3 takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Gemma 4 26B A4B
Gemma 4 26B A4B $0.0004, MiniMax M3 $0.0009 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only MiniMax M3 has the number, $0.56.
Directional only
At a glance
- Coding index
- Gemma 4 26B A4B not benchmarked, MiniMax M3 39.2 from 6 benchmarks.
- Widest category
- Agentic work, 36.0 points apart on average over 2 shared benchmarks.
- Widest single gap
- τ²-Bench Telecom: Gemma 4 26B A4B 43.6% against MiniMax M3 88.9%, 45.3 points apart on a 0 to 100 scale.
- Output price
- Gemma 4 26B A4B $0.60 against MiniMax M3 $1.20 per million output tokens, 2.0x apart.
- Context window
- Gemma 4 26B A4B takes 262K tokens, MiniMax M3 1.05M.
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 | Gemma 4 26B A4B | Bars | MiniMax M3 | Ties |
|---|---|---|---|---|
| Coding | 1 | 8 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 0 | 2 | ||
| Reasoning | 1 | 7 | ||
| Math | 3 | 0 | ||
| Knowledge | 1 | 3 | ||
| General | 12 | 12 | ||
| All shared boards | 18 | 34 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index18
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 70.9% | Only MiniMax M3 measured |
| LMArena WebDev | 1360 | 1482 | MiniMax M3 |
| AA Coding Index | 39.3 | 58.6 | MiniMax M3 |
| LiveCodeBench (Vals AI) | Not measured | 82.2% | Only MiniMax M3 measured |
| SWE-bench Verified (Vals AI) | Not measured | 75.0% | Only MiniMax M3 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 1.0% | Only MiniMax M3 measured |
| Vibe Code Bench (Vals AI) | Not measured | 47.6% | Only MiniMax M3 measured |
| Code migration (Vals AI) | Not measured | 19.9% | Only MiniMax M3 measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 53.6% | Only MiniMax M3 measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 33.3% | Only MiniMax M3 measured |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 9.2% | Only MiniMax M3 measured |
| SkillsBench (Vals AI) | Not measured | 51.5% | Only MiniMax M3 measured |
| CyberBench (Vals AI) | Not measured | 65.0% | Only MiniMax M3 measured |
| SciCode | 40.0% | 47.1% | MiniMax M3 |
| Terminal-Bench Hard (Artificial Analysis run) | 25.0% | 42.4% | MiniMax M3 |
| FrontierCode | Not measured | 14.7% | Only MiniMax M3 measured |
| WeirdML | 35.2% | Not measured | Only Gemma 4 26B A4B measured |
| ALE-Bench | 927.2 | 640.0 | Gemma 4 26B A4B |
| LMArena Text: coding | 1446 | 1470 | MiniMax M3 |
| LMArena Text: software and IT | 1454 | 1466 | MiniMax M3 |
| LMArena WebDev: React | 1351 | 1478 | MiniMax M3 |
| LMArena WebDev: HTML | 1362 | 1455 | MiniMax M3 |
| LMArena WebDev: from an image | Not measured | 1470 | Only MiniMax M3 measured |
| Neither model has a result on the other 16 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 16.7 | 29.2 | MiniMax M3 |
| Epoch Capabilities Index | 141.9 | 146.9 | MiniMax M3 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | Leader |
|---|---|---|---|
| AA Agentic Index | 2.9 | 29.5 | MiniMax M3 |
| Design Arena: agents building web apps | Not measured | 1200 | Only MiniMax M3 measured |
| Design Arena: agents building full-stack apps | Not measured | 1197 | Only MiniMax M3 measured |
| Design Arena: agents building mobile apps | Not measured | 1169 | Only MiniMax M3 measured |
| Design Arena: agents building Android apps | Not measured | 1159 | Only MiniMax M3 measured |
| Design Arena: agents building games | Not measured | 1159 | Only MiniMax M3 measured |
| Design Arena: HTML slides | Not measured | 1130 | Only MiniMax M3 measured |
| Design Arena: PowerPoint slides | Not measured | 1181 | Only MiniMax M3 measured |
| Design Arena: spreadsheets | Not measured | 1039 | Only MiniMax M3 measured |
| LiveBench research | Not measured | 68.0% | Only MiniMax M3 measured |
| LiveBench browser | Not measured | 8.0% | Only MiniMax M3 measured |
| LiveBench long agentic | Not measured | 12.4% | Only MiniMax M3 measured |
| Finance Agent (Vals AI) | Not measured | 48.3% | Only MiniMax M3 measured |
| Tax Agent Bench (Vals AI) | Not measured | 20.6% | Only MiniMax M3 measured |
| Legal Research Bench (Vals AI) | Not measured | 29.8% | Only MiniMax M3 measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 47.8% | Only MiniMax M3 measured |
| τ²-Bench Telecom | 43.6% | 88.9% | MiniMax M3 |
| APEX-Agents | Not measured | 37.7% | Only MiniMax M3 measured |
| OSWorld 2.0 | Not measured | 4.6% | Only MiniMax M3 measured |
| Vending-Bench 2 | Not measured | $2,158 | Only MiniMax M3 measured |
| Neither model has a result on the other 21 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index17
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 92.7% | Only MiniMax M3 measured |
| LiveBench reasoning | Not measured | 74.5% | Only MiniMax M3 measured |
| MMMU Pro (Vals AI) | Not measured | 81.2% | Only MiniMax M3 measured |
| GPQA Diamond (Artificial Analysis run) | 79.2% | 92.9% | MiniMax M3 |
| Humanity's Last Exam (Artificial Analysis run) | 19.3% | 39.0% | MiniMax M3 |
| GPQA Diamond (Epoch AI run) | 73.2% | 90.9% | MiniMax M3 |
| SimpleBench | Not measured | 45.8% | Only MiniMax M3 measured |
| Chess puzzles | Not measured | 13.0% | Only MiniMax M3 measured |
| Mystery game puzzles | Not measured | 8.0% | Only MiniMax M3 measured |
| LMArena Text: hard prompts | 1438 | 1447 | MiniMax M3 |
| LMArena Text: expert | 1447 | 1461 | MiniMax M3 |
| LMArena Text: hard prompts in English | 1448 | 1455 | MiniMax M3 |
| LMArena Vision: diagrams | 1264 | 1266 | MiniMax M3 |
| LMArena Vision: homework | 1293 | 1268 | Gemma 4 26B A4B |
| Neither model has a result on the other 8 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 82.2% | 71.1% | Gemma 4 26B A4B |
| ProofBench | Not measured | 18.0% | Only MiniMax M3 measured |
| LMArena Text: math | 1470 | 1429 | Gemma 4 26B A4B |
| LMArena Text: mathematical work | 1475 | 1448 | Gemma 4 26B A4B |
| Neither model has a result on the other 9 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index13
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 84.2% | Only MiniMax M3 measured |
| LegalBench (Vals AI) | Not measured | 85.4% | Only MiniMax M3 measured |
| CorpFin (Vals AI) | Not measured | 68.1% | Only MiniMax M3 measured |
| TaxEval (Vals AI) | Not measured | 72.7% | Only MiniMax M3 measured |
| MortgageTax (Vals AI) | Not measured | 68.4% | Only MiniMax M3 measured |
| MedCode (Vals AI) | Not measured | 46.3% | Only MiniMax M3 measured |
| LMArena Text: science | 1446 | 1452 | MiniMax M3 |
| LMArena Text: medicine and healthcare | 1428 | 1441 | MiniMax M3 |
| LMArena Text: legal and government | 1457 | 1437 | Gemma 4 26B A4B |
| LMArena Text: business and finance | 1429 | 1434 | MiniMax M3 |
| FACTS Benchmark Suite | 42.7% | Not measured | Only Gemma 4 26B A4B measured |
| Neither model has a result on the other 7 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1257 | Only MiniMax M3 measured |
| Design Arena: UI components | Not measured | 1243 | Only MiniMax M3 measured |
| Design Arena: game dev | Not measured | 1211 | Only MiniMax M3 measured |
| Design Arena: data viz | Not measured | 1236 | Only MiniMax M3 measured |
| Design Arena: 3D | Not measured | 1208 | Only MiniMax M3 measured |
| Design Arena: all code categories | Not measured | 1246 | Only MiniMax M3 measured |
| Design Arena: SVG | Not measured | 1160 | Only MiniMax M3 measured |
| Design Arena: ASCII art | Not measured | 1161 | Only MiniMax M3 measured |
| LiveBench design | Not measured | 66.2% | Only MiniMax M3 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index1212
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 76.2% | Only MiniMax M3 measured |
| Public Benefits Bench (Vals AI) | Not measured | 64.1% | Only MiniMax M3 measured |
| SAGE (Vals AI) | Not measured | 50.6% | Only MiniMax M3 measured |
| MedScribe (Vals AI) | Not measured | 87.3% | Only MiniMax M3 measured |
| Vals Multimodal Index | Not measured | 60.0% | Only MiniMax M3 measured |
| IFBench | 72.4% | 82.9% | MiniMax M3 |
| Long context reasoning (AA-LCR) | 65.7% | 83.0% | MiniMax M3 |
| DTBench | 74.9% | 78.9% | MiniMax M3 |
| LMCA | 29.7% | 33.7% | MiniMax M3 |
| Surface Evolver Bench | Not measured | 55.0% | Only MiniMax M3 measured |
| LMArena Text | 1434 | 1433 | Gemma 4 26B A4B |
| LMArena Text: instruction following | 1420 | 1433 | MiniMax M3 |
| LMArena Text: creative writing | 1400 | 1404 | MiniMax M3 |
| LMArena Vision | 1260 | 1251 | Gemma 4 26B A4B |
| LMArena Text: English | 1444 | 1443 | Gemma 4 26B A4B |
| LMArena Text: non-English | 1421 | 1420 | Gemma 4 26B A4B |
| LMArena Text: Chinese | 1495 | 1466 | Gemma 4 26B A4B |
| LMArena Text: German | Not measured | 1428 | Only MiniMax M3 measured |
| LMArena Text: Japanese | Not measured | 1380 | Only MiniMax M3 measured |
| LMArena Text: Korean | Not measured | 1373 | Only MiniMax M3 measured |
| LMArena Text: French | 1460 | 1444 | Gemma 4 26B A4B |
| LMArena Text: Spanish | 1415 | 1433 | MiniMax M3 |
| LMArena Text: Russian | 1434 | 1426 | Gemma 4 26B A4B |
| LMArena Text: Polish | Not measured | 1423 | Only MiniMax M3 measured |
| LMArena Text: multi-turn | 1441 | 1443 | MiniMax M3 |
| LMArena Text: longer queries | 1427 | 1446 | MiniMax M3 |
| LMArena Text: writing and language | 1405 | 1419 | MiniMax M3 |
| LMArena Text: entertainment and media | 1403 | 1403 | Gemma 4 26B A4B |
| LMArena Vision: English | 1256 | 1263 | MiniMax M3 |
| LMArena Vision: Chinese | 1312 | 1311 | Gemma 4 26B A4B |
| LMArena Vision: reading text | 1268 | 1259 | Gemma 4 26B A4B |
| LMArena Vision: humor | 1272 | 1235 | Gemma 4 26B A4B |
| LMArena Vision: captioning | 1297 | Not measured | Only Gemma 4 26B A4B measured |
| LMArena Vision: recognizing things | 1191 | 1197 | MiniMax M3 |
| LMArena Vision: creative writing | 1256 | 1234 | Gemma 4 26B A4B |
| LMArena Document | Not measured | 1435 | Only MiniMax M3 measured |
| Neither model has a result on the other 10 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Gemma 4 26B A4B | MiniMax M3 | 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
Gemma 4 26B A4B costs less
Repository review
50K input + 3K output tokens
Gemma 4 26B A4B costs less
Agent loop
200K input + 10K output tokens
Gemma 4 26B A4B 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 | Google Gemma 4 26B A4B | MiniMax M3 |
|---|---|---|
| Maker | MiniMax | |
| Weights | Open weights | Open weights |
| Released | April 3, 2026 | May 31, 2026 |
| Context window | 262K | 1.05M |
| Input $/Mtok | $0.15 | $0.30 |
| Output $/Mtok | $0.60 | $1.20 |
| Coding index | Not benchmarked | 39.2 |
| Independent measurements behind it | 1 | 6 |
| Cost per solved task | Not computable | $0.56 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Gemma 4 26B A4B better than MiniMax M3 for coding?
Only MiniMax M3 has a coding score, 39.2 on the Vibe Coding Index; Gemma 4 26B A4B has none, so the two cannot be ranked on coding. Pick Gemma 4 26B A4B for output tokens 2.0x cheaper ($0.60 vs $1.20 per million) and work measured by ALE-Bench or LMArena Vision: homework, where it leads.
Which is cheaper, Gemma 4 26B A4B or MiniMax M3?
Gemma 4 26B A4B costs $0.15 per million input tokens and $0.60 per million output tokens; MiniMax M3 costs $0.30 and $1.20. Cost per solved task is known for MiniMax M3 only.
Which has the bigger context window, Gemma 4 26B A4B or MiniMax M3?
MiniMax M3, with 1.05M tokens against 262K.
How many benchmarks is the Gemma 4 26B A4B vs MiniMax M3 comparison based on?
52 benchmarks have a result for both models as of October 22, 2026. Gemma 4 26B A4B's coding index rests on 1 independent benchmark and MiniMax M3's on 6. Where only one model has a result, the page says so instead of naming a winner.