Model comparison
Llama 2-70B vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.2 to 24.1 on the Noometry Index.
Last verified . 23 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Qwen3 235B-A22B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 235B-A22B leads 49.5 to 7.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0% for Llama 2-70B and 86.7% for Qwen3 235B-A22B.
Side by side
| Llama 2-70B | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 24.1 | 43.2 |
| Released | 2023-07-18 | 2025-04 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $2.80 |
| Results tracked | 35 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Llama 2-70B: 30.8 (#281), Qwen3 235B-A22B: 43.8 (#72)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Coding | 1079 | 1445 |
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
| WeirdML | — | 41% |
Agentic & Tool Use Not comparable
Llama 2-70B: —, Qwen3 235B-A22B: 33.7 (#48)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Vending-Bench 2 | — | -11.34 |
Reasoning Qwen3 235B-A22B leads
Llama 2-70B: 13.7 (#324), Qwen3 235B-A22B: 15.8 (#302)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1433 |
| DTBench | 41.6% | 80.3% |
| Epoch Capabilities Index | 113.59 | 143.87 |
| ForecastBench | 51.4 | 59.7 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Chess Puzzles | — | 12% |
| Mystery Game Puzzles | — | 9% |
| LMCA | — | 29.3% |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math Qwen3 235B-A22B leads
Llama 2-70B: 7.5 (#320), Qwen3 235B-A22B: 48.5 (#51)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0% | 86.7% |
| LMArena Math | 1091 | 1431 |
| MATH Level 5 | 3.3% | 68.9% |
| Omni-MATH | — | 71.8% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
| GSM8K | 69.6% | — |
Knowledge Qwen3 235B-A22B leads
Llama 2-70B: 7.4 (#305), Qwen3 235B-A22B: 49.5 (#67)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| GPQA Diamond | 26.3% | 80.1% |
| LMArena Expert | 1039 | 1460 |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Qwen3 235B-A22B leads
Llama 2-70B: 27.9 (#269), Qwen3 235B-A22B: 52.5 (#84)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1045 | 1410 |
| LMArena Chinese | 994 | 1481 |
| LMArena French | 1089 | 1445 |
| LMArena German | 1041 | 1432 |
| LMArena Japanese | 927 | 1400 |
| LMArena Korean | 964 | 1391 |
| LMArena Russian | 1083 | 1411 |
| LMArena Spanish | 1142 | 1429 |
Instruction Following Qwen3 235B-A22B leads
Llama 2-70B: 55.1 (#273), Qwen3 235B-A22B: 72.8 (#132)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1071 | 1408 |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Llama 2-70B: 32.2 (#265), Qwen3 235B-A22B: 46.1 (#25)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1062 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Qwen3 235B-A22B leads
Llama 2-70B: 32.5 (#274), Qwen3 235B-A22B: 59.8 (#103)
| Benchmark | Llama 2-70B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1115 | 1419 |
| LMArena Creative Writing | 1075 | 1384 |
| LMArena Multi-Turn | 1088 | 1432 |
| Short-Story Creative Writing | — | 83% |
| EQ-Bench Creative Writing | — | 1366 |
| WildBench | — | 86.6% |
Frequently asked questions
Is Llama 2-70B better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.2 to 24.1 on the Noometry Index.
Is Llama 2-70B or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 43.8 versus 30.8 in the Noometry coding category.
How many benchmarks do Llama 2-70B and Qwen3 235B-A22B share?
23 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Qwen3 235B-A22B has 49.