# DeepSeek LLM 67B vs DeepSeek V4 Flash

> DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 24.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-llm-67b-vs-deepseek-v4-flash
- Last updated: 2026-10-11
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and DeepSeek V4 Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Flash leads 60.3 to 8.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 94.4% for DeepSeek V4 Flash.

## Snapshot

| | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 24.9 | 53.6 |
| Rank | 347 | 35 |
| Context | — | 1M |
| Input $/M | — | $0.15 |
| Output $/M | — | $0.60 |
| Weights | Open | Open |

## Coding

- DeepSeek LLM 67B: 31.9 (#278)
- DeepSeek V4 Flash: 47.9 (#59)

| Benchmark | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Coding | 1096 | 1457 |
| FrontierCode | — | 18.8% |
| LMArena WebDev | — | 1582 |
| SciCode | — | 49.9% |
| WeirdML | — | 63% |
| ALE-Bench | — | 1,306 |

## Reasoning

- DeepSeek LLM 67B: 16.5 (#304)
- DeepSeek V4 Flash: 53.7 (#30)

| Benchmark | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| Chess Puzzles | 0% | 33% |
| LMArena Hard Prompts | 1070 | 1444 |
| Epoch Capabilities Index | 110.5 | 154.49 |
| ARC-AGI-2 | — | 61.4% |
| SimpleBench | — | 61.1% |
| Kagi LLM Benchmark | — | 52.2% |
| NYT Connections (extended) | — | 89.6% |
| ARC-AGI-1 | — | 89% |
| CritPt | — | 16.6% |
| Mystery Game Puzzles | — | 34% |
| DTBench | — | 90.9% |
| LMCA | — | 41.7% |

## Math

- DeepSeek LLM 67B: 8.7 (#324)
- DeepSeek V4 Flash: 60.3 (#37)

| Benchmark | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 94.4% |
| LMArena Math | 1108 | 1427 |
| FrontierMath (Tiers 1-3) | — | 57.5% |
| FrontierMath Tier 4 | — | 24.4% |
| MathArena Final-Answer Competitions | — | 76.5% |
| ProofBench | — | 56% |
| MATH Level 5 | 6.4% | — |

## Knowledge

- DeepSeek LLM 67B: 7.0 (#313)
- DeepSeek V4 Flash: 55.4 (#48)

| Benchmark | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| GPQA Diamond | 24.6% | 91% |
| SimpleQA Verified | — | 33.6% |
| LMArena Expert | — | 1441 |

## Multilingual

- DeepSeek LLM 67B: 29.4 (#267)
- DeepSeek V4 Flash: 53.0 (#72)

| Benchmark | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Non-English | 1073 | 1420 |
| LMArena Chinese | 1132 | 1468 |
| LMArena French | — | 1439 |
| LMArena German | — | 1418 |
| LMArena Japanese | — | 1406 |
| LMArena Korean | — | 1384 |
| LMArena Russian | — | 1428 |
| LMArena Spanish | — | 1436 |

## Instruction Following

- DeepSeek LLM 67B: 55.4 (#277)
- DeepSeek V4 Flash: 74.9 (#81)

| Benchmark | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Instruction Following | 1079 | 1421 |

## Long Context

- DeepSeek LLM 67B: 33.1 (#265)
- DeepSeek V4 Flash: 43.8 (#85)

| Benchmark | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Longer Query | 1092 | 1434 |

## Writing & Preference

- DeepSeek LLM 67B: 31.6 (#282)
- DeepSeek V4 Flash: 63.8 (#61)

| Benchmark | DeepSeek LLM 67B | DeepSeek V4 Flash |
|---|---|---|
| LMArena Text | 1105 | 1432 |
| LMArena Creative Writing | 1067 | 1403 |
| LMArena Multi-Turn | 1082 | 1449 |
| EQ-Bench Creative Writing | — | 1559 |

## FAQ

### Is DeepSeek LLM 67B better than DeepSeek V4 Flash?

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 24.9 on the Noometry Index.

### Is DeepSeek LLM 67B or DeepSeek V4 Flash better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 31.9 in the Noometry coding category.

### How many benchmarks do DeepSeek LLM 67B and DeepSeek V4 Flash share?

14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and DeepSeek V4 Flash has 41.