Model comparison
GPT-4 vs Nemotron 4 340b Instruct
Nemotron 4 340b Instruct is the stronger model overall, scoring 35.9 to 29.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
NVIDIASummary
- They share 17 benchmarks with published results for both. GPT-4 scores higher in 3 categories and Nemotron 4 340b Instruct in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Nemotron 4 340b Instruct leads 34.3 to 10.8.
- Nemotron 4 340b Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-4 | Nemotron 4 340b Instruct | |
|---|---|---|
| Provider | OpenAI | NVIDIA |
| Noometry Index | 29.1 | 35.9 |
| Released | 2023-03-14 | 2024-06-14 |
| Weights | Proprietary | Open |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $30 | — |
| Output $ / M tokens | $60 | — |
| Results tracked | 38 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 4 340b Instruct leads
GPT-4: 31.6 (#283), Nemotron 4 340b Instruct: 35.2 (#228)
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Coding | 1254 | 1209 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Nemotron 4 340b Instruct: —
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning Nemotron 4 340b Instruct leads
GPT-4: 17.8 (#289), Nemotron 4 340b Instruct: 23.4 (#178)
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1202 |
| 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 4 340b Instruct leads
GPT-4: 10.8 (#309), Nemotron 4 340b Instruct: 34.3 (#195)
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Math | 1269 | 1216 |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |
Knowledge Nemotron 4 340b Instruct leads
GPT-4: 18.4 (#282), Nemotron 4 340b Instruct: 32.0 (#216)
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Expert | 1211 | 1171 |
| GPQA Diamond | 35.7% | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual GPT-4 leads
GPT-4: 40.6 (#215), Nemotron 4 340b Instruct: 37.9 (#233)
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Non-English | 1246 | 1206 |
| LMArena Chinese | 1242 | 1214 |
| LMArena French | 1283 | 1197 |
| LMArena German | 1251 | 1192 |
| LMArena Japanese | 1209 | 1128 |
| LMArena Korean | 1184 | 1148 |
| LMArena Russian | 1251 | 1219 |
| LMArena Spanish | 1261 | 1195 |
Instruction Following GPT-4 leads
GPT-4: 65.3 (#222), Nemotron 4 340b Instruct: 63.1 (#232)
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Instruction Following | 1241 | 1203 |
Long Context Too close to call
GPT-4: 37.7 (#212), Nemotron 4 340b Instruct: 37.1 (#221)
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Longer Query | 1244 | 1225 |
Writing & Preference Nemotron 4 340b Instruct leads
GPT-4: 34.9 (#268), Nemotron 4 340b Instruct: 42.7 (#233)
| Benchmark | GPT-4 | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Text | 1263 | 1225 |
| LMArena Creative Writing | 1244 | 1203 |
| LMArena Multi-Turn | 1257 | 1214 |
| EQ-Bench Creative Writing | 752 | — |
Frequently asked questions
Is GPT-4 better than Nemotron 4 340b Instruct?
Nemotron 4 340b Instruct is the stronger model overall, scoring 35.9 to 29.1 on the Noometry Index.
Is GPT-4 or Nemotron 4 340b Instruct better for coding?
Nemotron 4 340b Instruct scores higher on coding benchmarks: 35.2 versus 31.6 in the Noometry coding category.
How many benchmarks do GPT-4 and Nemotron 4 340b Instruct share?
17 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Nemotron 4 340b Instruct has 17.