# DeepSeek LLM 67B vs Gemini 3.8 Flash

> Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 24.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-llm-67b-vs-gemini-3-8-flash
- Last updated: 2026-10-10
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Gemini 3.8 Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3.8 Flash leads 74.8 to 7.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 98.9% for Gemini 3.8 Flash.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| Provider | DeepSeek | Google |
| Noometry Index | 24.9 | 61.8 |
| Rank | 347 | 11 |
| Context | — | 1.05M |
| Input $/M | — | $0.75 |
| Output $/M | — | $3.75 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek LLM 67B: 31.9 (#278)
- Gemini 3.8 Flash: 59.2 (#15)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| LMArena Coding | 1096 | 1510 |
| DeepSWE | — | 73.8% |
| FrontierCode | — | 41.2% |
| CursorBench | — | 39.6% |
| LMArena WebDev | — | 1584 |
| FrontierSWE | — | 19.6% |
| SciCode | — | 56.6% |
| WeirdML | — | 84.8% |
| ALE-Bench | — | 1,270 |

## Agentic & Tool Use

- DeepSeek LLM 67B: —
- Gemini 3.8 Flash: 41.8 (#21)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| APEX-Agents | — | 64.3% |
| Remote Labor Index | — | 5.8% |
| GDP.pdf | — | 23.4% |
| Vending-Bench 2 | — | 5,094 |

## Reasoning

- DeepSeek LLM 67B: 16.5 (#304)
- Gemini 3.8 Flash: 76.9 (#5)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| Chess Puzzles | 0% | 61% |
| LMArena Hard Prompts | 1070 | 1508 |
| Epoch Capabilities Index | 110.5 | 156.71 |
| ARC-AGI-2 | — | 89.2% |
| NYT Connections (extended) | — | 97.4% |
| ARC-AGI-1 | — | 98.5% |
| CritPt | — | 18.3% |
| Mystery Game Puzzles | — | 47% |
| DTBench | — | 95.7% |
| LMCA | — | 52.9% |
| Surface Evolver Bench | — | 76.9% |

## Math

- DeepSeek LLM 67B: 8.7 (#324)
- Gemini 3.8 Flash: 65.3 (#28)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 98.9% |
| LMArena Math | 1108 | 1528 |
| FrontierMath (Tiers 1-3) | — | 68.4% |
| FrontierMath Tier 4 | — | 22% |
| ProofBench | — | 48% |
| MATH Level 5 | 6.4% | — |

## Knowledge

- DeepSeek LLM 67B: 7.0 (#313)
- Gemini 3.8 Flash: 74.8 (#2)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| GPQA Diamond | 24.6% | 95.4% |
| Humanity's Last Exam | — | 44.5% |
| SimpleQA Verified | — | 69.7% |
| LMArena Expert | — | 1524 |

## Multimodal

- DeepSeek LLM 67B: —
- Gemini 3.8 Flash: 40.7 (#45)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| LMArena Vision | — | 1314 |
| Blueprint-Bench 2 | — | 38.6% |
| Furniture Assembly | — | 31.7% |

## Multilingual

- DeepSeek LLM 67B: 29.4 (#267)
- Gemini 3.8 Flash: 58.0 (#5)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| LMArena Non-English | 1073 | 1491 |
| LMArena Chinese | 1132 | 1554 |
| LMArena French | — | 1498 |
| LMArena German | — | 1493 |
| LMArena Japanese | — | 1502 |
| LMArena Korean | — | 1459 |
| LMArena Russian | — | 1515 |
| LMArena Spanish | — | 1485 |

## Instruction Following

- DeepSeek LLM 67B: 55.4 (#277)
- Gemini 3.8 Flash: 78.0 (#13)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| LMArena Instruction Following | 1079 | 1490 |

## Long Context

- DeepSeek LLM 67B: 33.1 (#265)
- Gemini 3.8 Flash: 46.3 (#24)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| LMArena Longer Query | 1092 | 1508 |

## Writing & Preference

- DeepSeek LLM 67B: 31.6 (#282)
- Gemini 3.8 Flash: 72.2 (#15)

| Benchmark | DeepSeek LLM 67B | Gemini 3.8 Flash |
|---|---|---|
| LMArena Text | 1105 | 1499 |
| LMArena Creative Writing | 1067 | 1492 |
| LMArena Multi-Turn | 1082 | 1501 |
| EQ-Bench Creative Writing | — | 1748 |

## FAQ

### Is DeepSeek LLM 67B better than Gemini 3.8 Flash?

Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 24.9 on the Noometry Index.

### Is DeepSeek LLM 67B or Gemini 3.8 Flash better for coding?

Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 31.9 in the Noometry coding category.

### How many benchmarks do DeepSeek LLM 67B and Gemini 3.8 Flash share?

14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Gemini 3.8 Flash has 50.