# DeepSeek V4 Flash vs Yi-34B

> DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 27.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v4-flash-vs-yi-34b
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 8 categories and Yi-34B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek V4 Flash leads 55.5 to 7.5.
- The biggest single-benchmark swing is GPQA Diamond: 91% for DeepSeek V4 Flash and 14.7% for Yi-34B.

## Snapshot

| | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| Provider | DeepSeek | 01.AI |
| Noometry Index | 53.6 | 27.4 |
| Rank | 32 | 325 |
| Context | 1M | — |
| Input $/M | $0.15 | — |
| Output $/M | $0.60 | — |
| Weights | Open | Open |

## Coding

- DeepSeek V4 Flash: 47.6 (#56)
- Yi-34B: 31.8 (#269)

| Benchmark | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| LMArena Coding | 1457 | 1112 |
| FrontierCode | 18.8% | — |
| LMArena WebDev | 1581 | — |
| SciCode | 49.9% | — |
| WeirdML | 63% | — |
| ALE-Bench | 1,306 | — |

## Reasoning

- DeepSeek V4 Flash: 53.7 (#25)
- Yi-34B: 20.3 (#222)

| Benchmark | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| LMArena Hard Prompts | 1443 | 1104 |
| Epoch Capabilities Index | 154.51 | 117.28 |
| 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% | — |
| Chess Puzzles | 33% | — |
| Mystery Game Puzzles | 34% | — |
| DTBench | 90.9% | — |
| LMCA | 41.7% | — |
| BIG-Bench Hard | — | 71.7% |

## Math

- DeepSeek V4 Flash: 60.0 (#31)
- Yi-34B: 19.8 (#277)

| Benchmark | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| LMArena Math | 1427 | 1114 |
| FrontierMath (Tiers 1-3) | 57.5% | — |
| FrontierMath Tier 4 | 24.4% | — |
| MathArena Final-Answer Competitions | 76.5% | — |
| OTIS Mock AIME 2024-2025 | 94.4% | — |
| ProofBench | 56% | — |
| MATH Level 5 | — | 5.1% |
| GSM8K | — | 76% |

## Knowledge

- DeepSeek V4 Flash: 55.5 (#43)
- Yi-34B: 7.5 (#304)

| Benchmark | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| GPQA Diamond | 91% | 14.7% |
| LMArena Expert | 1441 | 1061 |
| SimpleQA Verified | 33.6% | — |
| MMLU | — | 76.3% |

## Multilingual

- DeepSeek V4 Flash: 53.2 (#71)
- Yi-34B: 29.9 (#259)

| Benchmark | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| LMArena Non-English | 1420 | 1079 |
| LMArena Chinese | 1467 | 1175 |
| LMArena French | 1438 | 1080 |
| LMArena German | 1418 | 1042 |
| LMArena Japanese | 1403 | 993 |
| LMArena Korean | 1383 | 959 |
| LMArena Russian | 1428 | 1050 |
| LMArena Spanish | 1437 | 1070 |

## Instruction Following

- DeepSeek V4 Flash: 75.1 (#76)
- Yi-34B: 56.4 (#269)

| Benchmark | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| LMArena Instruction Following | 1421 | 1091 |

## Long Context

- DeepSeek V4 Flash: 43.8 (#81)
- Yi-34B: 33.2 (#259)

| Benchmark | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| LMArena Longer Query | 1435 | 1094 |

## Writing & Preference

- DeepSeek V4 Flash: 64.0 (#59)
- Yi-34B: 34.3 (#268)

| Benchmark | DeepSeek V4 Flash | Yi-34B |
|---|---|---|
| LMArena Text | 1432 | 1129 |
| LMArena Creative Writing | 1403 | 1108 |
| LMArena Multi-Turn | 1449 | 1113 |
| EQ-Bench Creative Writing | 1559 | — |

## FAQ

### Is DeepSeek V4 Flash better than Yi-34B?

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 27.4 on the Noometry Index.

### Is DeepSeek V4 Flash or Yi-34B better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.6 versus 31.8 in the Noometry coding category.

### How many benchmarks do DeepSeek V4 Flash and Yi-34B share?

19 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Yi-34B has 23.