Model comparison
GPT-5 Mini vs Nemotron 3 Ultra
Nemotron 3 Ultra is the stronger model overall, scoring 44.7 to 41.7 on the Noometry Index. GPT-5 Mini costs 1.5× less per token, which makes it the better buy when Nemotron 3 Ultra's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
NVIDIASummary
- They share 28 benchmarks with published results for both. GPT-5 Mini scores higher in 3 categories and Nemotron 3 Ultra in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 34.4.
- The biggest single-benchmark swing is Chess Puzzles: 30% for GPT-5 Mini and 12% for Nemotron 3 Ultra.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $0.60 / $2.40 for Nemotron 3 Ultra.
- GPT-5 Mini accepts more context: 400K tokens versus 262K.
- Nemotron 3 Ultra has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Mini | Nemotron 3 Ultra | |
|---|---|---|
| Provider | OpenAI | NVIDIA |
| Noometry Index | 41.7 | 44.7 |
| Released | 2025-08-07 | 2026-06-04 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 128K |
| Input $ / M tokens | $0.25 | $0.60 |
| Output $ / M tokens | $2 | $2.40 |
| Results tracked | 60 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-5 Mini: 40.0 (#134), Nemotron 3 Ultra: 37.7 (#174)
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| SciCode | 39.2% | 40.3% |
| WeirdML | 52.7% | 43.5% |
| LMArena Coding | 1406 | 1468 |
| SWE-bench Verified | 64.7% | — |
| FrontierCode | — | 13.6% |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use Not comparable
GPT-5 Mini: 31.0 (#64), Nemotron 3 Ultra: —
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning Nemotron 3 Ultra leads
GPT-5 Mini: 24.1 (#153), Nemotron 3 Ultra: 30.3 (#78)
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 0% | 3.1% |
| Chess Puzzles | 30% | 12% |
| LMArena Hard Prompts | 1380 | 1452 |
| Mystery Game Puzzles | 10% | 20% |
| DTBench | 80.5% | 90.1% |
| LMCA | 34.2% | 36.9% |
| Epoch Capabilities Index | 145.51 | 146.2 |
| ARC-AGI-2 | 4.4% | — |
| Kagi LLM Benchmark | 70.3% | — |
| ARC-AGI-1 | 54.3% | — |
| EnigmaEval | 8.2% | — |
| ForecastBench | 61 | — |
Math GPT-5 Mini leads
GPT-5 Mini: 46.3 (#59), Nemotron 3 Ultra: 34.4 (#160)
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 86.7% |
| ProofBench | 9% | 2% |
| LMArena Math | 1377 | 1456 |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| Omni-MATH | 72.2% | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge Nemotron 3 Ultra leads
GPT-5 Mini: 45.5 (#80), Nemotron 3 Ultra: 52.5 (#56)
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 75% | 85.4% |
| LMArena Expert | 1377 | 1472 |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| MMLU-Pro | 83.5% | — |
| Confabulations | 13.3% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |
Multimodal Not comparable
GPT-5 Mini: 35.0 (#80), Nemotron 3 Ultra: —
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |
Multilingual Nemotron 3 Ultra leads
GPT-5 Mini: 49.1 (#131), Nemotron 3 Ultra: 53.7 (#61)
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1363 | 1427 |
| LMArena Chinese | 1387 | 1498 |
| LMArena French | 1385 | 1475 |
| LMArena German | 1366 | 1471 |
| LMArena Korean | 1306 | 1387 |
| LMArena Russian | 1362 | 1415 |
| LMArena Spanish | 1355 | 1452 |
| LMArena Japanese | 1342 | — |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.4 (#43), Nemotron 3 Ultra: 74.9 (#84)
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1357 | 1417 |
| IFEval | 92.7% | — |
Long Context Nemotron 3 Ultra leads
GPT-5 Mini: 41.9 (#127), Nemotron 3 Ultra: 43.9 (#77)
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1355 | 1437 |
| Fiction.LiveBench | 69.4% | — |
Writing & Preference Nemotron 3 Ultra leads
GPT-5 Mini: 55.4 (#143), Nemotron 3 Ultra: 66.3 (#35)
| Benchmark | GPT-5 Mini | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1373 | 1444 |
| LMArena Creative Writing | 1325 | 1392 |
| EQ-Bench Creative Writing | 1313 | 1692 |
| LMArena Multi-Turn | 1363 | 1412 |
| Short-Story Creative Writing | 83.1% | — |
| WildBench | 85.5% | — |
Frequently asked questions
Is GPT-5 Mini better than Nemotron 3 Ultra?
Nemotron 3 Ultra is the stronger model overall, scoring 44.7 to 41.7 on the Noometry Index. GPT-5 Mini costs 1.5× less per token, which makes it the better buy when Nemotron 3 Ultra's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Mini or Nemotron 3 Ultra?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Nemotron 3 Ultra lists at $0.60 and $2.40.
Is GPT-5 Mini or Nemotron 3 Ultra better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.0 versus 37.7 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 262K.
How many benchmarks do GPT-5 Mini and Nemotron 3 Ultra share?
28 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Nemotron 3 Ultra has 29.