Gemini 4 Argon vs MiniMax-M3
Published test results, prices, limits and features, side by side. Share the link to show someone exactly this comparison.
Gemini 4 ArgonGoogle
MiniMax-M3MiniMax
At a glance
Capability index
- Gemini 4 Argon164.6
- MiniMax-M3147.1
Gemini 4 Argon is 17.4 points higher.
Price per 1M tokens
- Gemini 4 Argon$4.00
- MiniMax-M3$0.525
MiniMax-M3 is 7.6 times cheaper.
Context window
- Gemini 4 Argon1M tokens
- MiniMax-M31M tokens
They take in about the same.
Released
- Gemini 4 ArgonOct 13, 2026
- MiniMax-M3Jun 1, 2026
Gemini 4 Argon is 4 months newer.
Where each scores higher
On the 10 tests both have taken, each on its own scale. A gap of a point or two can sit within a test’s margin of error.
Gemini 4 Argon
- Arena text+83 points
- Arena WebDev+196 points
- Arena coding+65 points
- Mock AIME 2024–2025+28.9 points
- Arena math+98 points
- Arena creative writing+111 points
- Arena instruction following+93 points
- GPQA Diamond+5.5 points
- Chess puzzles+46.0 points
- Arena hard prompts+86 points
MiniMax-M3
No shared test where it scores higher.
12 more tests have results for only one of them; the table below lists every result.
Everything side by side
| Measure | Gemini 4 ArgonGoogle | MiniMax-M3MiniMax |
|---|---|---|
| Model | ||
| Maker | MiniMax | |
| Sold here by | MiniMax (minimax.io) | |
| API model ID | gemini-4-argon | MiniMax-M3 |
| Released | Oct 13, 2026 | Jun 1, 2026 |
| Knowledge cutoff | Mar 2026 | Jan 2025 |
| Weights | Closed: hosted access only | Open: downloadable |
| Developer’s country | United States | China |
| Hosts selling it | 24including Google directly | 60including MiniMax directly |
| Test results | ||
| Capability index | 164.6 (best of these models)#5 of 277 | 147.1#70 of 277 |
| Arena text | 1525 (best of these models)#1 of 415 | 1441#96 of 415 |
| LiveBench overall | Not tested | 67.3%#65 of 70 |
| Arena WebDev | 1678 (best of these models)#9 of 142 | 1482#62 of 142 |
| Arena coding | 1561 (best of these models)#1 of 410 | 1496#87 of 410 |
| LiveBench coding | Not tested | 68.2%#68 of 70 |
| Mock AIME 2024–2025 | 100.0% (best of these models)#1 of 303 | 71.1%#139 of 303 |
| FrontierMath Tiers 1–3 | 86.0%#10 of 117 | Not tested |
| FrontierMath Tier 4 | 78.0%#15 of 73 | Not tested |
| Arena math | 1529 (best of these models)#1 of 400 | 1430#108 of 400 |
| LiveBench mathematics | Not tested | 77.0%#69 of 70 |
| SimpleQA Verified | 76.0%#1 of 90 | Not tested |
| CursorBench | 52.9%#6 of 71 | Not tested |
| LiveBench agentic coding | Not tested | 40.7%#64 of 70 |
| Arena creative writing | 1516 (best of these models)#2 of 413 | 1406#103 of 413 |
| Arena instruction following | 1528 (best of these models)#1 of 415 | 1435#89 of 415 |
| LiveBench language | Not tested | 76.8%#46 of 70 |
| LiveBench instructions | Not tested | 57.5%#64 of 70 |
| LiveBench data analysis | Not tested | 76.2%#36 of 70 |
| GPQA Diamond | 96.4% (best of these models)#1 of 323 | 90.9%#35 of 323 |
| Chess puzzles | 59.0% (best of these models)#9 of 243 | 13.0%#134 of 243 |
| Arena hard prompts | 1550 (best of these models)#1 of 415 | 1464#90 of 415 |
| LiveBench reasoning | Not tested | 74.5%#65 of 70 |
| Price per million tokens | ||
| Input | $2.00 | $0.30 (best of these models) |
| Output | $10.00 | $1.20 (best of these models) |
| Blended, 3 input to 1 output | $4.00 | $0.525 (best of these models) |
| Cached input | $0.10 | $0.06 (best of these models) |
| Audio input | $2.00 | Not reported |
| Long requests | No separate price reported | $0.60 in, $2.40 outabove 512K tokens |
| The maker’s own price | This listing | This listing |
| Limits | ||
| Context window | 1M tokens | 1M tokens |
| Max input | 1M tokens (best of these models) | 512K tokens |
| Max output | 32.8K tokens | 512K tokens (best of these models) |
| Features | ||
| Reads | audio, images, PDFs, text and video | images, text and video |
| Produces | text | text |
| Reasoning | Yeseffort low, medium, high | Yeseffort low, medium, high; can be switched off; thinking budget can be set |
| Tool calling | Yes | Yes |
| Structured output | Yes, such as JSON | Yes, such as JSON |
| File attachments | Yes | Yes |
| Temperature setting | Supported | Supported |
Bold marks the better value in each row. Prices are each model’s own list price, or the middle price across hosts when the maker does not sell it, blended as three input tokens for every output token. Test results as published by LiveBench (Apache-2.0), Epoch AI (CC BY 4.0), Cursor CursorBench, compiled by Epoch AI (CC BY 4.0), Arena (CC BY 4.0); model details and prices from models.dev (MIT). Confirm pricing with the provider before use.