Model comparison
Claude 2 vs DeepSeek V4 Flash
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 25.0 on the Noometry Index.
Last verified . 4 shared benchmarks.
DeepSeekSummary
- They share 4 benchmarks with published results for both. Claude 2 scores higher in 0 categories and DeepSeek V4 Flash in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4 Flash leads 60.3 to 9.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.5% for Claude 2 and 94.4% for DeepSeek V4 Flash.
- DeepSeek V4 Flash has downloadable open weights; the other is API-only.
Side by side
| Claude 2 | DeepSeek V4 Flash | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 25.0 | 53.6 |
| Released | 2023-07-11 | 2026-04-24 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 393K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 8 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Claude 2: —, DeepSeek V4 Flash: 47.9 (#59)
| Benchmark | Claude 2 | DeepSeek V4 Flash |
|---|---|---|
| FrontierCode | — | 18.8% |
| LMArena WebDev | — | 1582 |
| SciCode | — | 49.9% |
| WeirdML | — | 63% |
| LMArena Coding | — | 1457 |
| ALE-Bench | — | 1,306 |
| HumanEval+ | 61.6% | — |
Reasoning DeepSeek V4 Flash leads
Claude 2: 21.7 (#216), DeepSeek V4 Flash: 53.7 (#30)
| Benchmark | Claude 2 | DeepSeek V4 Flash |
|---|---|---|
| DTBench | 51.9% | 90.9% |
| Epoch Capabilities Index | 120.13 | 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% |
| Chess Puzzles | — | 33% |
| LMArena Hard Prompts | — | 1444 |
| Mystery Game Puzzles | — | 34% |
| LMCA | — | 41.7% |
Math DeepSeek V4 Flash leads
Claude 2: 9.3 (#320), DeepSeek V4 Flash: 60.3 (#37)
| Benchmark | Claude 2 | DeepSeek V4 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 94.4% |
| FrontierMath (Tiers 1-3) | — | 57.5% |
| FrontierMath Tier 4 | — | 24.4% |
| MathArena Final-Answer Competitions | — | 76.5% |
| ProofBench | — | 56% |
| LMArena Math | — | 1427 |
| MATH Level 5 | 11.7% | — |
Knowledge DeepSeek V4 Flash leads
Claude 2: 16.9 (#287), DeepSeek V4 Flash: 55.4 (#48)
| Benchmark | Claude 2 | DeepSeek V4 Flash |
|---|---|---|
| GPQA Diamond | 34.7% | 91% |
| SimpleQA Verified | — | 33.6% |
| LMArena Expert | — | 1441 |
| MMLU | 78.5% | — |
| TriviaQA | 87.5% | — |
Multilingual Not comparable
Claude 2: —, DeepSeek V4 Flash: 53.0 (#72)
| Benchmark | Claude 2 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Non-English | — | 1420 |
| LMArena Chinese | — | 1468 |
| LMArena French | — | 1439 |
| LMArena German | — | 1418 |
| LMArena Japanese | — | 1406 |
| LMArena Korean | — | 1384 |
| LMArena Russian | — | 1428 |
| LMArena Spanish | — | 1436 |
Instruction Following Not comparable
Claude 2: —, DeepSeek V4 Flash: 74.9 (#81)
| Benchmark | Claude 2 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Instruction Following | — | 1421 |
Long Context Not comparable
Claude 2: —, DeepSeek V4 Flash: 43.8 (#85)
| Benchmark | Claude 2 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Longer Query | — | 1434 |
Writing & Preference Not comparable
Claude 2: —, DeepSeek V4 Flash: 63.8 (#61)
| Benchmark | Claude 2 | DeepSeek V4 Flash |
|---|---|---|
| LMArena Text | — | 1432 |
| LMArena Creative Writing | — | 1403 |
| EQ-Bench Creative Writing | — | 1559 |
| LMArena Multi-Turn | — | 1449 |
Frequently asked questions
Is Claude 2 better than DeepSeek V4 Flash?
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 25.0 on the Noometry Index.
How many benchmarks do Claude 2 and DeepSeek V4 Flash share?
4 benchmarks have published results for both models. Claude 2 has 8 scored results on Noometry and DeepSeek V4 Flash has 41.