Model comparison
DeepSeek-V2.5 (Sep 2024) vs GPT-5 Nano
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 33.5 on the Noometry Index.
Last verified . 16 shared benchmarks.
DeepSeekSummary
- They share 16 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 4 categories and GPT-5 Nano in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V2.5 (Sep 2024) leads 49.8 to 39.1.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 37.6 | 33.5 |
| Released | 2024-09-06 | 2025-08-07 |
| Weights | Open | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.05 |
| Output $ / M tokens | — | $0.40 |
| Results tracked | 22 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Nano leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), GPT-5 Nano: 33.6 (#254)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Coding | 1309 | 1351 |
| SWE-bench Verified (bash only) | — | 34.8% |
| Aider Polyglot | 17.8% | — |
| WeirdML | — | 38.1% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 718.67 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, GPT-5 Nano: 25.8 (#106)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| Terminal-Bench | — | 21.8% |
| Berkeley Function Calling Leaderboard | — | 51.5% |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), GPT-5 Nano: 16.3 (#306)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1328 |
| ARC-AGI-2 | — | 2.6% |
| Kagi LLM Benchmark | — | 62.2% |
| ARC-AGI-1 | — | 20.7% |
| Chess Puzzles | — | 27% |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 62.7% |
| LMCA | — | 7.9% |
| Epoch Capabilities Index | — | 139.38 |
| ForecastBench | — | 59.1 |
Math DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), GPT-5 Nano: 29.4 (#241)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Math | 1288 | 1317 |
| FrontierMath (Tiers 1-3) | — | 20% |
| FrontierMath Tier 4 | — | 2.4% |
| OTIS Mock AIME 2024-2025 | — | 81.1% |
| ProofBench | — | 12% |
| Omni-MATH | — | 54.6% |
| MATH Level 5 | — | 95.2% |
| FrontierMath (Feb 2025 set) | — | 8.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GPT-5 Nano leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), GPT-5 Nano: 35.9 (#178)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Expert | 1266 | 1321 |
| GPQA Diamond | — | 69.4% |
| SimpleQA Verified | — | 11.7% |
| MMLU-Pro | — | 77.8% |
| Vectara Hallucination Rate | — | 10.5% |
| GPQA (HELM) | — | 67.9% |
Multimodal Not comparable
DeepSeek-V2.5 (Sep 2024): —, GPT-5 Nano: 31.3 (#108)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Vision | — | 1159 |
| VPCT | — | 37.2% |
Multilingual GPT-5 Nano leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), GPT-5 Nano: 45.3 (#172)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Non-English | 1273 | 1313 |
| LMArena Chinese | 1318 | 1356 |
| LMArena German | 1258 | 1327 |
| LMArena Japanese | 1228 | 1226 |
| LMArena Korean | 1209 | 1269 |
| LMArena Russian | 1289 | 1296 |
| LMArena Spanish | 1248 | 1360 |
| LMArena French | 1289 | — |
Instruction Following GPT-5 Nano leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), GPT-5 Nano: 75.0 (#79)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Instruction Following | 1280 | 1306 |
| IFEval | — | 93.2% |
Long Context DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), GPT-5 Nano: 31.3 (#281)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Longer Query | 1301 | 1312 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), GPT-5 Nano: 39.1 (#249)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5 Nano |
|---|---|---|
| LMArena Text | 1294 | 1320 |
| LMArena Creative Writing | 1285 | 1249 |
| LMArena Multi-Turn | 1297 | 1311 |
| EQ-Bench Creative Writing | — | 705 |
| WildBench | — | 80.6% |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than GPT-5 Nano?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 33.5 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or GPT-5 Nano better for coding?
GPT-5 Nano scores higher on coding benchmarks: 33.6 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and GPT-5 Nano share?
16 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and GPT-5 Nano has 49.