Model comparison
Llama 3.1 Nemotron Ultra 253b v1 vs Qwen2-72B
Llama 3.1 Nemotron Ultra 253b v1 is the stronger model overall, scoring 36.7 to 30.0 on the Noometry Index.
Last verified . 10 shared benchmarks.
NVIDIA
Alibaba (Qwen)Summary
- They share 10 benchmarks with published results for both. Llama 3.1 Nemotron Ultra 253b v1 scores higher in 7 categories and Qwen2-72B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Llama 3.1 Nemotron Ultra 253b v1 leads 52.2 to 40.8.
Side by side
| Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B | |
|---|---|---|
| Provider | NVIDIA | Alibaba (Qwen) |
| Noometry Index | 36.7 | 30.0 |
| Released | — | 2024-06-07 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 11 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3.1 Nemotron Ultra 253b v1 leads
Llama 3.1 Nemotron Ultra 253b v1: 38.4 (#177), Qwen2-72B: 29.1 (#310)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1312 | 1196 |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
Agentic & Tool Use Qwen2-72B leads
Llama 3.1 Nemotron Ultra 253b v1: 15.7 (#149), Qwen2-72B: 17.0 (#146)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 10% | — |
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Llama 3.1 Nemotron Ultra 253b v1 leads
Llama 3.1 Nemotron Ultra 253b v1: 26.3 (#134), Qwen2-72B: 23.2 (#181)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1316 | 1191 |
| Epoch Capabilities Index | — | 125.28 |
Math Llama 3.1 Nemotron Ultra 253b v1 leads
Llama 3.1 Nemotron Ultra 253b v1: 37.5 (#152), Qwen2-72B: 30.2 (#236)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| LMArena Math | 1360 | 1235 |
| MATH Level 5 | — | 39.1% |
Knowledge Not comparable
Llama 3.1 Nemotron Ultra 253b v1: —, Qwen2-72B: 21.2 (#275)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| GPQA Diamond | — | 40.8% |
| LMArena Expert | — | 1171 |
| MMLU | — | 82.4% |
Multilingual Llama 3.1 Nemotron Ultra 253b v1 leads
Llama 3.1 Nemotron Ultra 253b v1: 43.1 (#187), Qwen2-72B: 35.9 (#244)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1282 | 1176 |
| LMArena Russian | 1284 | 1169 |
| LMArena Chinese | — | 1240 |
| LMArena French | — | 1170 |
| LMArena German | — | 1151 |
| LMArena Japanese | — | 1111 |
| LMArena Korean | — | 1083 |
| LMArena Spanish | — | 1169 |
Instruction Following Llama 3.1 Nemotron Ultra 253b v1 leads
Llama 3.1 Nemotron Ultra 253b v1: 69.0 (#178), Qwen2-72B: 61.7 (#241)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1308 | 1181 |
Long Context Llama 3.1 Nemotron Ultra 253b v1 leads
Llama 3.1 Nemotron Ultra 253b v1: 39.5 (#177), Qwen2-72B: 36.1 (#235)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1299 | 1192 |
Writing & Preference Llama 3.1 Nemotron Ultra 253b v1 leads
Llama 3.1 Nemotron Ultra 253b v1: 52.2 (#175), Qwen2-72B: 40.8 (#241)
| Benchmark | Llama 3.1 Nemotron Ultra 253b v1 | Qwen2-72B |
|---|---|---|
| LMArena Text | 1320 | 1203 |
| LMArena Creative Writing | 1314 | 1181 |
| LMArena Multi-Turn | 1317 | 1196 |
Frequently asked questions
Is Llama 3.1 Nemotron Ultra 253b v1 better than Qwen2-72B?
Llama 3.1 Nemotron Ultra 253b v1 is the stronger model overall, scoring 36.7 to 30.0 on the Noometry Index.
Is Llama 3.1 Nemotron Ultra 253b v1 or Qwen2-72B better for coding?
Llama 3.1 Nemotron Ultra 253b v1 scores higher on coding benchmarks: 38.4 versus 29.1 in the Noometry coding category.
How many benchmarks do Llama 3.1 Nemotron Ultra 253b v1 and Qwen2-72B share?
10 benchmarks have published results for both models. Llama 3.1 Nemotron Ultra 253b v1 has 11 scored results on Noometry and Qwen2-72B has 26.