Model comparison
Gemini 3.8 Flash vs Nemotron 3 Ultra
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 1.6× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Google
NVIDIASummary
- They share 30 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 9 categories and Nemotron 3 Ultra in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 31.0.
- The biggest single-benchmark swing is Chess Puzzles: 61% for Gemini 3.8 Flash and 12% for Nemotron 3 Ultra.
- Nemotron 3 Ultra is cheaper at $0.50 / $2.20 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.8 Flash.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 262K.
- Nemotron 3 Ultra has downloadable open weights; the other is API-only.
Side by side
| Gemini 3.8 Flash | Nemotron 3 Ultra | |
|---|---|---|
| Provider | NVIDIA | |
| Noometry Index | 61.8 | 42.5 |
| Released | 2026-09-02 | 2026-06-04 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.75 | $0.50 |
| Output $ / M tokens | $3.75 | $2.20 |
| Results tracked | 50 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3.8 Flash leads
Gemini 3.8 Flash: 59.2 (#15), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| FrontierCode | 41.2% | 13.6% |
| SciCode | 56.6% | 40.3% |
| WeirdML | 84.8% | 43.5% |
| LMArena Coding | 1510 | 1468 |
| DeepSWE | 73.8% | — |
| CursorBench | 39.6% | — |
| LMArena WebDev | 1584 | — |
| FrontierSWE | 19.6% | — |
| ALE-Bench | 1,270 | — |
Agentic & Tool Use Gemini 3.8 Flash leads
Gemini 3.8 Flash: 41.8 (#21), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| APEX-Agents | 64.3% | 22.7% |
| Remote Labor Index | 5.8% | — |
| GDP.pdf | 23.4% | — |
| Vending-Bench 2 | 5,094 | — |
Reasoning Gemini 3.8 Flash leads
Gemini 3.8 Flash: 76.9 (#5), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 18.3% | 3.1% |
| Chess Puzzles | 61% | 12% |
| LMArena Hard Prompts | 1508 | 1452 |
| Mystery Game Puzzles | 47% | 20% |
| DTBench | 95.7% | 90.1% |
| LMCA | 52.9% | 36.9% |
| Epoch Capabilities Index | 156.71 | 146.17 |
| ARC-AGI-2 | 89.2% | — |
| NYT Connections (extended) | 97.4% | — |
| ARC-AGI-1 | 98.5% | — |
| Surface Evolver Bench | 76.9% | — |
Math Gemini 3.8 Flash leads
Gemini 3.8 Flash: 65.3 (#28), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 86.7% |
| ProofBench | 48% | 2% |
| LMArena Math | 1528 | 1457 |
| FrontierMath (Tiers 1-3) | 68.4% | — |
| FrontierMath Tier 4 | 22% | — |
Knowledge Gemini 3.8 Flash leads
Gemini 3.8 Flash: 74.8 (#2), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 95.4% | 85.4% |
| LMArena Expert | 1524 | 1472 |
| Humanity's Last Exam | 44.5% | — |
| SimpleQA Verified | 69.7% | — |
Multimodal Not comparable
Gemini 3.8 Flash: 40.7 (#45), Nemotron 3 Ultra: —
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| LMArena Vision | 1314 | — |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 31.7% | — |
Multilingual Gemini 3.8 Flash leads
Gemini 3.8 Flash: 58.0 (#5), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1491 | 1427 |
| LMArena Chinese | 1554 | 1497 |
| LMArena French | 1498 | 1472 |
| LMArena German | 1493 | 1471 |
| LMArena Korean | 1459 | 1386 |
| LMArena Russian | 1515 | 1417 |
| LMArena Spanish | 1485 | 1454 |
| LMArena Japanese | 1502 | — |
Instruction Following Gemini 3.8 Flash leads
Gemini 3.8 Flash: 78.0 (#13), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1490 | 1418 |
Long Context Gemini 3.8 Flash leads
Gemini 3.8 Flash: 46.3 (#24), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1508 | 1437 |
Writing & Preference Gemini 3.8 Flash leads
Gemini 3.8 Flash: 72.2 (#15), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | Gemini 3.8 Flash | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1499 | 1445 |
| LMArena Creative Writing | 1492 | 1398 |
| EQ-Bench Creative Writing | 1748 | 1692 |
| LMArena Multi-Turn | 1501 | 1411 |
Frequently asked questions
Is Gemini 3.8 Flash better than Nemotron 3 Ultra?
Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 42.5 on the Noometry Index. Nemotron 3 Ultra costs 1.6× less per token, which makes it the better buy when Gemini 3.8 Flash's lead doesn't matter for your workload.
Which is cheaper, Gemini 3.8 Flash or Nemotron 3 Ultra?
Nemotron 3 Ultra is cheaper. It lists at $0.50 per million input tokens and $2.20 per million output tokens; Gemini 3.8 Flash lists at $0.75 and $3.75.
Is Gemini 3.8 Flash or Nemotron 3 Ultra better for coding?
Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
Gemini 3.8 Flash does, with 1.05M tokens against 262K.
How many benchmarks do Gemini 3.8 Flash and Nemotron 3 Ultra share?
30 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and Nemotron 3 Ultra has 30.