Model comparison
DeepSeek-V2.5 (Sep 2024) vs Qwen3-4B
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.0 to 31.5 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- The widest gap is in reasoning, where DeepSeek-V2.5 (Sep 2024) leads 24.5 to 18.5.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Qwen3-4B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 37.0 | 31.5 |
| Released | 2024-09-06 | 2025-04-29 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
DeepSeek-V2.5 (Sep 2024): 31.3 (#274), Qwen3-4B: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| LMArena Coding | 1310 | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Qwen3-4B: 27.3 (#94)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 35.7% |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 24.5 (#145), Qwen3-4B: 18.5 (#264)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1289 | — |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 33.3 (#177), Qwen3-4B: 28.5 (#228)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| MathArena Final-Answer Competitions | — | 38.5% |
| OTIS Mock AIME 2024-2025 | — | 52.2% |
| LMArena Math | 1289 | — |
Knowledge DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 34.2 (#190), Qwen3-4B: 33.0 (#198)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| GPQA Diamond | — | 52.3% |
| Vectara Hallucination Rate | — | 5.7% |
| LMArena Expert | 1266 | — |
Multilingual Not comparable
DeepSeek-V2.5 (Sep 2024): 42.7 (#188), Qwen3-4B: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1273 | — |
| LMArena Chinese | 1317 | — |
| LMArena French | 1288 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
| LMArena Russian | 1289 | — |
| LMArena Spanish | 1248 | — |
Instruction Following Not comparable
DeepSeek-V2.5 (Sep 2024): 67.7 (#189), Qwen3-4B: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| LMArena Instruction Following | 1280 | — |
Long Context Not comparable
DeepSeek-V2.5 (Sep 2024): 39.5 (#169), Qwen3-4B: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| LMArena Longer Query | 1301 | — |
Writing & Preference Not comparable
DeepSeek-V2.5 (Sep 2024): 50.0 (#182), Qwen3-4B: —
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Qwen3-4B |
|---|---|---|
| LMArena Text | 1294 | — |
| LMArena Creative Writing | 1285 | — |
| LMArena Multi-Turn | 1297 | — |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Qwen3-4B?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.0 to 31.5 on the Noometry Index.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Qwen3-4B share?
0 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Qwen3-4B has 6.