# Gemini 3.8 Flash vs Grok 4.20 (Non-Reasoning)

> Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 48.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-8-flash-vs-grok-4-20
- Last updated: 2026-10-10
- Shared benchmarks: 37

## Summary

- They share 37 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 10 categories and Grok 4.20 (Non-Reasoning) in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 52.3.
- The biggest single-benchmark swing is SimpleQA Verified: 69.7% for Gemini 3.8 Flash and 30.2% for Grok 4.20 (Non-Reasoning).
- Both cost about the same: $0.75 input and $3.75 output per million tokens.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 1M.

## Snapshot

| | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Provider | Google | xAI |
| Noometry Index | 61.8 | 48.6 |
| Rank | 11 | 54 |
| Context | 1.05M | 1M |
| Input $/M | $0.75 | $1.25 |
| Output $/M | $3.75 | $2.50 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.8 Flash: 59.2 (#15)
- Grok 4.20 (Non-Reasoning): 42.1 (#112)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena WebDev | 1584 | 1375 |
| WeirdML | 84.8% | 52.3% |
| LMArena Coding | 1510 | 1459 |
| ALE-Bench | 1,270 | 1,150 |
| DeepSWE | 73.8% | — |
| FrontierCode | 41.2% | — |
| CursorBench | 39.6% | — |
| FrontierSWE | 19.6% | — |
| SciCode | 56.6% | — |

## Agentic & Tool Use

- Gemini 3.8 Flash: 41.8 (#21)
- Grok 4.20 (Non-Reasoning): 34.4 (#46)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Vending-Bench 2 | 5,094 | 4,663 |
| Terminal-Bench | — | 57.3% |
| APEX-Agents | 64.3% | — |
| Remote Labor Index | 5.8% | — |
| τ²-bench Banking | — | 18% |
| GDP.pdf | 23.4% | — |
| LMArena Search | — | 1189 |

## Reasoning

- Gemini 3.8 Flash: 76.9 (#5)
- Grok 4.20 (Non-Reasoning): 52.3 (#32)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 89.2% | 65.1% |
| NYT Connections (extended) | 97.4% | 85.4% |
| ARC-AGI-1 | 98.5% | 89.5% |
| Chess Puzzles | 61% | 24% |
| LMArena Hard Prompts | 1508 | 1451 |
| DTBench | 95.7% | 90.1% |
| LMCA | 52.9% | 38.7% |
| Epoch Capabilities Index | 156.71 | 151.98 |
| Kagi LLM Benchmark | — | 75% |
| CritPt | 18.3% | — |
| Thematic Generalization | — | 63.8% |
| Mystery Game Puzzles | 47% | — |
| Surface Evolver Bench | 76.9% | — |
| ForecastBench | — | 61.4 |

## Math

- Gemini 3.8 Flash: 65.3 (#28)
- Grok 4.20 (Non-Reasoning): 48.2 (#65)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.4% | 44.9% |
| FrontierMath Tier 4 | 22% | 17.1% |
| OTIS Mock AIME 2024-2025 | 98.9% | 92.2% |
| ProofBench | 48% | 14% |
| LMArena Math | 1528 | 1455 |

## Knowledge

- Gemini 3.8 Flash: 74.8 (#2)
- Grok 4.20 (Non-Reasoning): 52.8 (#60)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 95.4% | 89.3% |
| SimpleQA Verified | 69.7% | 30.2% |
| LMArena Expert | 1524 | 1439 |
| Humanity's Last Exam | 44.5% | — |

## Multimodal

- Gemini 3.8 Flash: 40.7 (#45)
- Grok 4.20 (Non-Reasoning): 33.3 (#98)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Vision | 1314 | 1263 |
| Blueprint-Bench 2 | 38.6% | 0% |
| Furniture Assembly | 31.7% | — |
| LMArena Document | — | 1416 |

## Multilingual

- Gemini 3.8 Flash: 58.0 (#5)
- Grok 4.20 (Non-Reasoning): 54.5 (#40)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1491 | 1441 |
| LMArena Chinese | 1554 | 1481 |
| LMArena French | 1498 | 1476 |
| LMArena German | 1493 | 1465 |
| LMArena Japanese | 1502 | 1449 |
| LMArena Korean | 1459 | 1417 |
| LMArena Russian | 1515 | 1458 |
| LMArena Spanish | 1485 | 1443 |

## Instruction Following

- Gemini 3.8 Flash: 78.0 (#13)
- Grok 4.20 (Non-Reasoning): 74.8 (#83)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1490 | 1420 |

## Long Context

- Gemini 3.8 Flash: 46.3 (#24)
- Grok 4.20 (Non-Reasoning): 45.5 (#34)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Longer Query | 1508 | 1437 |
| CL-bench | — | 22.2% |
| CL-bench Life | — | 11.9% |

## Writing & Preference

- Gemini 3.8 Flash: 72.2 (#15)
- Grok 4.20 (Non-Reasoning): 65.7 (#44)

| Benchmark | Gemini 3.8 Flash | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1499 | 1451 |
| LMArena Creative Writing | 1492 | 1438 |
| EQ-Bench Creative Writing | 1748 | 1574 |
| LMArena Multi-Turn | 1501 | 1456 |

## FAQ

### Is Gemini 3.8 Flash better than Grok 4.20 (Non-Reasoning)?

Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 48.6 on the Noometry Index.

### Which is cheaper, Gemini 3.8 Flash or Grok 4.20 (Non-Reasoning)?

Gemini 3.8 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.

### Is Gemini 3.8 Flash or Grok 4.20 (Non-Reasoning) better for coding?

Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 42.1 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.8 Flash does, with 1.05M tokens against 1M.

### How many benchmarks do Gemini 3.8 Flash and Grok 4.20 (Non-Reasoning) share?

37 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.