Model comparison
DeepSeek LLM 67B vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 24.8 on the Noometry Index.
Last verified . 14 shared benchmarks.
DeepSeekSummary
- They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and GPT-5.4 mini in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5.4 mini leads 51.4 to 7.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.8% for DeepSeek LLM 67B and 88.9% for GPT-5.4 mini.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek LLM 67B | GPT-5.4 mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 24.8 | 44.8 |
| Released | 2023-11-29 | 2026-03-17 |
| Weights | Open | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.75 |
| Output $ / M tokens | — | $4.50 |
| Results tracked | 15 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
DeepSeek LLM 67B: 31.3 (#273), GPT-5.4 mini: 45.0 (#67)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| LMArena Coding | 1096 | 1438 |
| FrontierCode | — | 27% |
| LMArena WebDev | — | 1397 |
| SciCode | — | 49.9% |
| WeirdML | — | 60.3% |
| ALE-Bench | — | 1,189 |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, GPT-5.4 mini: 29.6 (#74)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| DeepResearch Bench | — | 36.3% |
Reasoning GPT-5.4 mini leads
DeepSeek LLM 67B: 15.9 (#300), GPT-5.4 mini: 30.3 (#77)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| Chess Puzzles | 0% | 24% |
| LMArena Hard Prompts | 1070 | 1424 |
| Epoch Capabilities Index | 110.54 | 148.83 |
| ARC-AGI-2 | — | 18.9% |
| Kagi LLM Benchmark | — | 37.9% |
| NYT Connections (extended) | — | 61.8% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | — | 10% |
| Thematic Generalization | — | 61.7% |
| Mystery Game Puzzles | — | 11% |
| DTBench | — | 80% |
| LMCA | — | 40.8% |
| ForecastBench | — | 57 |
Math GPT-5.4 mini leads
DeepSeek LLM 67B: 8.1 (#318), GPT-5.4 mini: 45.4 (#63)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 88.9% |
| LMArena Math | 1108 | 1418 |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 9.8% |
| ProofBench | — | 21% |
| MATH Level 5 | 6.4% | — |
| FrontierMath (Feb 2025 set) | — | 28.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GPT-5.4 mini leads
DeepSeek LLM 67B: 7.1 (#308), GPT-5.4 mini: 51.4 (#62)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 24.6% | 86.9% |
| SimpleQA Verified | — | 29.4% |
| Vectara Hallucination Rate | — | 5.5% |
| LMArena Expert | — | 1434 |
Multimodal Not comparable
DeepSeek LLM 67B: —, GPT-5.4 mini: 39.0 (#51)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | — | 1245 |
Multilingual GPT-5.4 mini leads
DeepSeek LLM 67B: 29.5 (#262), GPT-5.4 mini: 52.1 (#94)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1073 | 1405 |
| LMArena Chinese | 1132 | 1446 |
| LMArena French | — | 1440 |
| LMArena German | — | 1407 |
| LMArena Japanese | — | 1375 |
| LMArena Korean | — | 1369 |
| LMArena Russian | — | 1417 |
| LMArena Spanish | — | 1403 |
Instruction Following GPT-5.4 mini leads
DeepSeek LLM 67B: 55.7 (#272), GPT-5.4 mini: 74.3 (#99)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1079 | 1405 |
Long Context GPT-5.4 mini leads
DeepSeek LLM 67B: 33.1 (#260), GPT-5.4 mini: 42.9 (#107)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1092 | 1407 |
Writing & Preference GPT-5.4 mini leads
DeepSeek LLM 67B: 31.8 (#277), GPT-5.4 mini: 64.2 (#56)
| Benchmark | DeepSeek LLM 67B | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1105 | 1411 |
| LMArena Creative Writing | 1067 | 1370 |
| LMArena Multi-Turn | 1082 | 1429 |
| EQ-Bench Creative Writing | — | 1665 |
Frequently asked questions
Is DeepSeek LLM 67B better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 24.8 on the Noometry Index.
Is DeepSeek LLM 67B or GPT-5.4 mini better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 31.3 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and GPT-5.4 mini share?
14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and GPT-5.4 mini has 46.