Model comparison
DeepSeek-V2.5 (Sep 2024) vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 37.6 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) 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 51.5 to 34.8.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 37.6 | 45.0 |
| Released | 2024-09-06 | 2026-03-17 |
| Weights | Open | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.75 |
| Output $ / M tokens | — | $4.50 |
| Results tracked | 22 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), GPT-5.4 mini: 45.2 (#72)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Coding | 1309 | 1438 |
| FrontierCode | — | 27% |
| Aider Polyglot | 17.8% | — |
| LMArena WebDev | — | 1397 |
| SciCode | — | 49.9% |
| WeirdML | — | 60.3% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 1,189 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, GPT-5.4 mini: 29.9 (#81)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| DeepResearch Bench | — | 36.3% |
Reasoning GPT-5.4 mini leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), GPT-5.4 mini: 30.4 (#85)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1424 |
| ARC-AGI-2 | — | 18.9% |
| Kagi LLM Benchmark | — | 37.9% |
| NYT Connections (extended) | — | 61.8% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | — | 10% |
| Chess Puzzles | — | 24% |
| Thematic Generalization | — | 61.7% |
| Mystery Game Puzzles | — | 11% |
| DTBench | — | 80% |
| LMCA | — | 40.8% |
| Epoch Capabilities Index | — | 148.84 |
| ForecastBench | — | 57 |
Math GPT-5.4 mini leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), GPT-5.4 mini: 45.5 (#75)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Math | 1288 | 1419 |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 9.8% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 21% |
| FrontierMath (Feb 2025 set) | — | 28.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GPT-5.4 mini leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), GPT-5.4 mini: 51.5 (#67)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Expert | 1266 | 1435 |
| GPQA Diamond | — | 86.9% |
| SimpleQA Verified | — | 29.4% |
| Vectara Hallucination Rate | — | 5.5% |
Multimodal Not comparable
DeepSeek-V2.5 (Sep 2024): —, GPT-5.4 mini: 39.7 (#56)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | — | 1245 |
Multilingual GPT-5.4 mini leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), GPT-5.4 mini: 51.9 (#96)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1273 | 1405 |
| LMArena Chinese | 1318 | 1446 |
| LMArena French | 1289 | 1440 |
| LMArena German | 1258 | 1409 |
| LMArena Japanese | 1228 | 1374 |
| LMArena Korean | 1209 | 1368 |
| LMArena Russian | 1289 | 1417 |
| LMArena Spanish | 1248 | 1405 |
Instruction Following GPT-5.4 mini leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), GPT-5.4 mini: 74.1 (#102)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1280 | 1405 |
Long Context GPT-5.4 mini leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), GPT-5.4 mini: 43.0 (#112)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1301 | 1407 |
Writing & Preference GPT-5.4 mini leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), GPT-5.4 mini: 64.0 (#58)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1294 | 1412 |
| LMArena Creative Writing | 1285 | 1370 |
| LMArena Multi-Turn | 1297 | 1429 |
| EQ-Bench Creative Writing | — | 1665 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or GPT-5.4 mini better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and GPT-5.4 mini share?
17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and GPT-5.4 mini has 46.