Model comparison
GPT-4 vs Nemotron 3.5 Lightning
Nemotron 3.5 Lightning is the stronger model overall, scoring 40.0 to 29.1 on the Noometry Index.
Last verified . 18 shared benchmarks.
NVIDIASummary
- They share 18 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Nemotron 3.5 Lightning in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Nemotron 3.5 Lightning leads 37.5 to 10.8.
- Nemotron 3.5 Lightning is cheaper at $0.05 / $0.20 per million input/output tokens, against $30 / $60 for GPT-4.
- Nemotron 3.5 Lightning accepts more context: 262K tokens versus 8K.
- Nemotron 3.5 Lightning has downloadable open weights; the other is API-only.
Side by side
| GPT-4 | Nemotron 3.5 Lightning | |
|---|---|---|
| Provider | OpenAI | NVIDIA |
| Noometry Index | 29.1 | 40.0 |
| Released | 2023-03-14 | 2026-08-11 |
| Weights | Proprietary | Open |
| Context window | 8K | 262K |
| Max output | 8K | 262K |
| Input $ / M tokens | $30 | $0.05 |
| Output $ / M tokens | $60 | $0.20 |
| Results tracked | 38 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3.5 Lightning leads
GPT-4: 31.6 (#283), Nemotron 3.5 Lightning: 40.4 (#141)
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| LMArena Coding | 1254 | 1375 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Nemotron 3.5 Lightning: —
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning Nemotron 3.5 Lightning leads
GPT-4: 17.8 (#289), Nemotron 3.5 Lightning: 26.8 (#127)
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1337 |
| Chess Puzzles | 4% | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 62.7% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| Epoch Capabilities Index | 125.89 | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math Nemotron 3.5 Lightning leads
GPT-4: 10.8 (#309), Nemotron 3.5 Lightning: 37.5 (#155)
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| LMArena Math | 1269 | 1359 |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |
Knowledge Nemotron 3.5 Lightning leads
GPT-4: 18.4 (#282), Nemotron 3.5 Lightning: 37.5 (#154)
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| LMArena Expert | 1211 | 1356 |
| GPQA Diamond | 35.7% | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual Nemotron 3.5 Lightning leads
GPT-4: 40.6 (#215), Nemotron 3.5 Lightning: 44.0 (#180)
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| LMArena Non-English | 1246 | 1295 |
| LMArena Chinese | 1242 | 1359 |
| LMArena French | 1283 | 1366 |
| LMArena German | 1251 | 1282 |
| LMArena Japanese | 1209 | 1206 |
| LMArena Korean | 1184 | 1238 |
| LMArena Russian | 1251 | 1253 |
| LMArena Spanish | 1261 | 1345 |
Instruction Following Nemotron 3.5 Lightning leads
GPT-4: 65.3 (#222), Nemotron 3.5 Lightning: 69.6 (#170)
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| LMArena Instruction Following | 1241 | 1318 |
Long Context Nemotron 3.5 Lightning leads
GPT-4: 37.7 (#212), Nemotron 3.5 Lightning: 39.9 (#165)
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| LMArena Longer Query | 1244 | 1314 |
Writing & Preference Nemotron 3.5 Lightning leads
GPT-4: 34.9 (#268), Nemotron 3.5 Lightning: 48.5 (#201)
| Benchmark | GPT-4 | Nemotron 3.5 Lightning |
|---|---|---|
| LMArena Text | 1263 | 1327 |
| LMArena Creative Writing | 1244 | 1254 |
| EQ-Bench Creative Writing | 752 | 1280 |
| LMArena Multi-Turn | 1257 | 1328 |
Frequently asked questions
Is GPT-4 better than Nemotron 3.5 Lightning?
Nemotron 3.5 Lightning is the stronger model overall, scoring 40.0 to 29.1 on the Noometry Index.
Which is cheaper, GPT-4 or Nemotron 3.5 Lightning?
Nemotron 3.5 Lightning is cheaper. It lists at $0.05 per million input tokens and $0.20 per million output tokens; GPT-4 lists at $30 and $60.
Is GPT-4 or Nemotron 3.5 Lightning better for coding?
Nemotron 3.5 Lightning scores higher on coding benchmarks: 40.4 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3.5 Lightning does, with 262K tokens against 8K.
How many benchmarks do GPT-4 and Nemotron 3.5 Lightning share?
18 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Nemotron 3.5 Lightning has 18.