Model comparison
Deepseek Coder v2 vs GPT-4.1
Deepseek Coder v2 and GPT-4.1 score almost the same on the Noometry Index (35.9 vs 35.9), so choose on price, context window or the category you care about most.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 3 categories and GPT-4.1 in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4.1 leads 57.6 to 38.2.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Deepseek Coder v2 | GPT-4.1 | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 35.9 | 35.9 |
| Released | 2024-06-17 | 2025-04-14 |
| Weights | Open | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 33K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $8 |
| Results tracked | 24 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 38.1 (#183), GPT-4.1: 34.4 (#238)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Coding | 1251 | 1391 |
| SWE-bench Verified | — | 48.5% |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| WeirdML | — | 39% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| CadEval | — | 42% |
| ALE-Bench | — | 558.1 |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Agentic & Tool Use Not comparable
Deepseek Coder v2: —, GPT-4.1: 34.7 (#43)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 54% |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), GPT-4.1: 11.7 (#339)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1384 |
| ARC-AGI-2 | — | 0.4% |
| SimpleBench | — | 27% |
| Kagi LLM Benchmark | — | 52.3% |
| ARC-AGI-1 | — | 5.5% |
| Chess Puzzles | — | 6% |
| EnigmaEval | — | 2.2% |
| DTBench | — | 68.3% |
| LMCA | — | 25.6% |
| Epoch Capabilities Index | — | 136.78 |
| ForecastBench | — | 61.5 |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), GPT-4.1: 22.3 (#280)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Math | 1241 | 1370 |
| FrontierMath (Tiers 1-3) | — | 6% |
| OTIS Mock AIME 2024-2025 | — | 38.3% |
| Omni-MATH | — | 47.1% |
| MATH Level 5 | — | 83% |
| FrontierMath (Feb 2025 set) | — | 5.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
| GSM8K | 94.5% | — |
Knowledge GPT-4.1 leads
Deepseek Coder v2: 32.3 (#212), GPT-4.1: 37.1 (#160)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Expert | 1181 | 1364 |
| GPQA Diamond | — | 66.9% |
| Humanity's Last Exam | — | 5.4% |
| SimpleQA Verified | — | 31.1% |
| MMLU-Pro | — | 81.1% |
| Vectara Hallucination Rate | — | 5.6% |
| GPQA (HELM) | — | 65.9% |
| ARC (AI2) Challenge | 64.3% | — |
Multimodal Not comparable
Deepseek Coder v2: —, GPT-4.1: 38.2 (#67)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Vision | — | 1211 |
| GeoBench | — | 72% |
Multilingual GPT-4.1 leads
Deepseek Coder v2: 36.3 (#240), GPT-4.1: 49.4 (#133)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1182 | 1370 |
| LMArena Chinese | 1201 | 1382 |
| LMArena French | 1185 | 1382 |
| LMArena German | 1164 | 1381 |
| LMArena Japanese | 1126 | 1319 |
| LMArena Korean | 1104 | 1339 |
| LMArena Russian | 1188 | 1377 |
| LMArena Spanish | 1153 | 1376 |
Instruction Following GPT-4.1 leads
Deepseek Coder v2: 61.7 (#242), GPT-4.1: 71.3 (#153)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Instruction Following | 1180 | 1367 |
| IFEval | — | 83.8% |
Long Context GPT-4.1 leads
Deepseek Coder v2: 37.0 (#224), GPT-4.1: 40.0 (#163)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1219 | 1385 |
| Fiction.LiveBench | — | 63.9% |
Writing & Preference GPT-4.1 leads
Deepseek Coder v2: 38.2 (#253), GPT-4.1: 57.6 (#125)
| Benchmark | Deepseek Coder v2 | GPT-4.1 |
|---|---|---|
| LMArena Text | 1191 | 1383 |
| LMArena Creative Writing | 1120 | 1363 |
| LMArena Multi-Turn | 1177 | 1398 |
| EQ-Bench Creative Writing | — | 1420 |
| WildBench | — | 85.4% |
Frequently asked questions
Is Deepseek Coder v2 better than GPT-4.1?
Deepseek Coder v2 and GPT-4.1 score almost the same on the Noometry Index (35.9 vs 35.9), so choose on price, context window or the category you care about most.
Is Deepseek Coder v2 or GPT-4.1 better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 34.4 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and GPT-4.1 share?
17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and GPT-4.1 has 52.