Gemini 3.7 Flash vs Grok 4.6
Published test results, prices, limits and features, side by side. Share the link to show someone exactly this comparison.
Gemini 3.7 FlashGoogle
Grok 4.6xAI
At a glance
Capability index
- Gemini 3.7 Flash157.3
- Grok 4.6156.4
Gemini 3.7 Flash is 0.8 points higher.
Price per 1M tokens
- Gemini 3.7 Flash$1.50
- Grok 4.6$3.00
Gemini 3.7 Flash is 2 times cheaper.
Context window
- Gemini 3.7 Flash1M tokens
- Grok 4.6500K tokens
Gemini 3.7 Flash takes in 2.1 times as much.
Released
- Gemini 3.7 FlashAug 13, 2026
- Grok 4.6Aug 12, 2026
Gemini 3.7 Flash is 1 day newer.
Where each scores higher
On the 22 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 3.7 Flash
- LiveBench overall+0.8 points
- Arena text+34 points
- LiveBench coding+2.1 points
- Arena coding+11 points
- LiveBench mathematics+0.9 points
- FrontierMath Tiers 1–3+5.6 points
- FrontierMath Tier 4+4.9 points
- Arena math+56 points
- SimpleQA Verified+19.9 points
- LiveBench agentic coding+1.3 points
- LiveBench language+1.8 points
- LiveBench instructions+8.1 points
- Arena creative writing+49 points
- Arena instruction following+34 points
- GPQA Diamond+0.8 points
- Chess puzzles+7.0 points
- Arena hard prompts+29 points
Grok 4.6
- Arena WebDev+28 points
- Mock AIME 2024–2025+2.0 points
- DeepSWE+2.0 points
- LiveBench data analysis+5.9 points
- LiveBench reasoning+2.7 points
1 more test has results for only one of them; the table below lists every result.
Everything side by side
| Measure | Gemini 3.7 FlashGoogle | Grok 4.6xAI |
|---|---|---|
| Model | ||
| Maker | xAI | |
| Sold here by | xAI | |
| API model ID | gemini-3.7-flash | grok-4.6 |
| Released | Aug 13, 2026 | Aug 12, 2026 |
| Knowledge cutoff | Mar 2026 | Feb 1, 2026 |
| Weights | Closed: hosted access only | Closed: hosted access only |
| Developer’s country | United States | United States |
| Hosts selling it | 29including Google directly | 35including xAI directly |
| Test results | ||
| Capability index | 157.3 (best of these models)#16 of 274 | 156.4#21 of 274 |
| LiveBench overall | 78.8% (best of these models)#16 of 66 | 78.0%#18 of 66 |
| Arena text | 1488 (best of these models)#17 of 413 | 1454#73 of 413 |
| LiveBench coding | 78.9% (best of these models)#29 of 66 | 76.8%#42 of 66 |
| Arena WebDev | 1592#27 of 138 | 1620 (best of these models)#23 of 138 |
| Arena coding | 1518 (best of these models)#41 of 408 | 1507#64 of 408 |
| LiveBench mathematics | 93.5% (best of these models)#20 of 66 | 92.6%#25 of 66 |
| Mock AIME 2024–2025 | 97.2%#31 of 297 | 99.2% (best of these models)#15 of 297 |
| FrontierMath Tiers 1–3 | 71.6% (best of these models)#23 of 114 | 66.0%#28 of 114 |
| FrontierMath Tier 4 | 36.6% (best of these models)#30 of 70 | 31.7%#33 of 70 |
| Arena math | 1503 (best of these models)#12 of 396 | 1448#78 of 396 |
| SimpleQA Verified | 69.2% (best of these models)#9 of 86 | 49.3%#27 of 86 |
| LiveBench agentic coding | 58.3% (best of these models)#19 of 66 | 57.0%#22 of 66 |
| DeepSWE | 65.5%#25 of 69 | 67.5% (best of these models)#20 of 69 |
| CursorBench | Not tested | 41.4%#24 of 62 |
| LiveBench language | 85.5% (best of these models)#13 of 66 | 83.7%#19 of 66 |
| LiveBench instructions | 79.9% (best of these models)#2 of 66 | 71.9%#20 of 66 |
| Arena creative writing | 1493 (best of these models)#5 of 411 | 1445#52 of 411 |
| Arena instruction following | 1485 (best of these models)#17 of 413 | 1451#63 of 413 |
| LiveBench data analysis | 68.0%#55 of 66 | 73.9% (best of these models)#40 of 66 |
| LiveBench reasoning | 87.8%#25 of 66 | 90.5% (best of these models)#12 of 66 |
| GPQA Diamond | 94.8% (best of these models)#5 of 319 | 94.0%#11 of 319 |
| Chess puzzles | 47.0% (best of these models)#14 of 227 | 40.0%#24 of 227 |
| Arena hard prompts | 1508 (best of these models)#19 of 413 | 1479#67 of 413 |
| Price per million tokens | ||
| Input | $0.75 (best of these models) | $2.00 |
| Output | $3.75 (best of these models) | $6.00 |
| Blended, 3 input to 1 output | $1.50 (best of these models) | $3.00 |
| Cached input | $0.075 (best of these models) | $0.50 |
| Audio input | $0.75 | Not reported |
| Long requests | No separate price reported | $4.00 in, $12.00 outabove 200K tokens |
| The maker’s own price | This listing | This listing |
| Limits | ||
| Context window | 1M tokens (best of these models) | 500K tokens |
| Max input | 936K tokens (best of these models) | 372K tokens |
| Max output | 65.5K tokens | 500K tokens (best of these models) |
| Features | ||
| Reads | audio, images, PDFs, text and video | images, PDFs and text |
| Produces | text | text |
| Reasoning | Yeseffort low, medium, high | Yeseffort low, medium, high, xhigh |
| 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), Datacurve DeepSWE, compiled by 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.