Model comparison
DeepSeek LLM 67B vs GPT-4.1 mini
GPT-4.1 mini is the stronger model overall, scoring 33.6 to 24.9 on the Noometry Index.
Last verified . 15 shared benchmarks.
DeepSeekSummary
- They share 15 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 3 categories and GPT-4.1 mini in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-4.1 mini leads 34.7 to 7.0.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 87.3% for GPT-4.1 mini.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek LLM 67B | GPT-4.1 mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 24.9 | 33.6 |
| Released | 2023-11-29 | 2025-04-14 |
| Weights | Open | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 33K |
| Input $ / M tokens | — | $0.40 |
| Output $ / M tokens | — | $1.60 |
| Results tracked | 15 | 47 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.9 (#278), GPT-4.1 mini: 30.6 (#293)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| LMArena Coding | 1096 | 1367 |
| SWE-bench Verified (bash only) | — | 23.9% |
| Aider Polyglot | — | 32.4% |
| SciCode | — | 40.4% |
| WeirdML | — | 37.6% |
| BigCodeBench Instruct | — | 48.9% |
| CadEval | — | 16% |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, GPT-4.1 mini: 33.3 (#55)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 50.5% |
Reasoning DeepSeek LLM 67B leads
DeepSeek LLM 67B: 16.5 (#304), GPT-4.1 mini: 10.8 (#340)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| Chess Puzzles | 0% | 7% |
| LMArena Hard Prompts | 1070 | 1349 |
| Epoch Capabilities Index | 110.5 | 135.01 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 48.6% |
| ARC-AGI-1 | — | 3.5% |
| CritPt | — | 0% |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 68.8% |
| LMCA | — | 21.1% |
Math GPT-4.1 mini leads
DeepSeek LLM 67B: 8.7 (#324), GPT-4.1 mini: 24.1 (#270)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 44.7% |
| LMArena Math | 1108 | 1343 |
| MATH Level 5 | 6.4% | 87.3% |
| FrontierMath (Tiers 1-3) | — | 6.7% |
| Omni-MATH | — | 49.1% |
| FrontierMath (Feb 2025 set) | — | 4.5% |
Knowledge GPT-4.1 mini leads
DeepSeek LLM 67B: 7.0 (#313), GPT-4.1 mini: 34.7 (#194)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| GPQA Diamond | 24.6% | 65.8% |
| SimpleQA Verified | — | 12.7% |
| MMLU-Pro | — | 78.3% |
| GPQA (HELM) | — | 61.4% |
| LMArena Expert | — | 1338 |
Multimodal Not comparable
DeepSeek LLM 67B: —, GPT-4.1 mini: 35.8 (#82)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| LMArena Vision | — | 1181 |
Multilingual GPT-4.1 mini leads
DeepSeek LLM 67B: 29.4 (#267), GPT-4.1 mini: 45.7 (#166)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| LMArena Non-English | 1073 | 1318 |
| LMArena Chinese | 1132 | 1329 |
| LMArena French | — | 1358 |
| LMArena German | — | 1351 |
| LMArena Japanese | — | 1290 |
| LMArena Korean | — | 1298 |
| LMArena Russian | — | 1324 |
| LMArena Spanish | — | 1319 |
Instruction Following GPT-4.1 mini leads
DeepSeek LLM 67B: 55.4 (#277), GPT-4.1 mini: 73.7 (#118)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| LMArena Instruction Following | 1079 | 1333 |
| IFEval | — | 90.4% |
Long Context DeepSeek LLM 67B leads
DeepSeek LLM 67B: 33.1 (#265), GPT-4.1 mini: 31.8 (#275)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| LMArena Longer Query | 1092 | 1344 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference GPT-4.1 mini leads
DeepSeek LLM 67B: 31.6 (#282), GPT-4.1 mini: 48.6 (#199)
| Benchmark | DeepSeek LLM 67B | GPT-4.1 mini |
|---|---|---|
| LMArena Text | 1105 | 1340 |
| LMArena Creative Writing | 1067 | 1300 |
| LMArena Multi-Turn | 1082 | 1354 |
| EQ-Bench Creative Writing | — | 1147 |
| WildBench | — | 83.8% |
Frequently asked questions
Is DeepSeek LLM 67B better than GPT-4.1 mini?
GPT-4.1 mini is the stronger model overall, scoring 33.6 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or GPT-4.1 mini better for coding?
DeepSeek LLM 67B scores higher on coding benchmarks: 31.9 versus 30.6 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and GPT-4.1 mini share?
15 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and GPT-4.1 mini has 47.