Model comparison
Llama 2-13B vs Qwen3 235B-A22B
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 29.6 on the Noometry Index.
Last verified . 20 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Llama 2-13B scores higher in 0 categories and Qwen3 235B-A22B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3 235B-A22B leads 59.6 to 29.8.
- The biggest single-benchmark swing is DTBench: 42.2% for Llama 2-13B and 80.3% for Qwen3 235B-A22B.
Side by side
| Llama 2-13B | Qwen3 235B-A22B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 29.6 | 43.5 |
| 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 | 32 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 235B-A22B leads
Llama 2-13B: 30.9 (#291), Qwen3 235B-A22B: 44.3 (#75)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Coding | 1062 | 1445 |
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
| WeirdML | — | 41% |
Agentic & Tool Use Not comparable
Llama 2-13B: —, Qwen3 235B-A22B: 33.9 (#51)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Vending-Bench 2 | — | -11.34 |
Reasoning Qwen3 235B-A22B leads
Llama 2-13B: 12.8 (#337), Qwen3 235B-A22B: 15.7 (#311)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| Chess Puzzles | 0% | 12% |
| LMArena Hard Prompts | 1051 | 1433 |
| DTBench | 42.2% | 80.3% |
| Epoch Capabilities Index | 106.17 | 143.85 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Mystery Game Puzzles | — | 9% |
| LMCA | — | 29.3% |
| BIG-Bench Hard | 58.2% | — |
| ForecastBench | — | 59.7 |
| HellaSwag | 80.7% | — |
| LAMBADA | 76.5% | — |
| PIQA | 80.8% | — |
| WinoGrande | 72.8% | — |
Math Qwen3 235B-A22B leads
Llama 2-13B: 31.1 (#229), Qwen3 235B-A22B: 50.4 (#57)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Math | 1065 | 1432 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| Omni-MATH | — | 71.8% |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
| GSM8K | 36.9% | — |
Knowledge Qwen3 235B-A22B leads
Llama 2-13B: 28.1 (#249), Qwen3 235B-A22B: 49.6 (#73)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Expert | 1030 | 1463 |
| GPQA Diamond | — | 80.1% |
| SimpleQA Verified | — | 40.4% |
| MMLU-Pro | — | 84.4% |
| Confabulations | — | 15.6% |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 72.7% |
| ARC (AI2) Challenge | 60.3% | — |
| BoolQ | 82.4% | — |
| MMLU | 55.6% | — |
| OpenBookQA | 57% | — |
| TriviaQA | 79.6% | — |
Multimodal Not comparable
Llama 2-13B: —, Qwen3 235B-A22B: —
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| ScienceQA | 55.8% | — |
Multilingual Qwen3 235B-A22B leads
Llama 2-13B: 26.5 (#279), Qwen3 235B-A22B: 52.3 (#89)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1024 | 1409 |
| LMArena Chinese | 1001 | 1481 |
| LMArena French | 1044 | 1445 |
| LMArena German | 1009 | 1433 |
| LMArena Japanese | 894 | 1399 |
| LMArena Korean | 953 | 1391 |
| LMArena Russian | 1055 | 1411 |
| LMArena Spanish | 1087 | 1430 |
Instruction Following Qwen3 235B-A22B leads
Llama 2-13B: 53.3 (#287), Qwen3 235B-A22B: 72.6 (#136)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Instruction Following | 1045 | 1408 |
| IFEval | — | 83.5% |
Long Context Qwen3 235B-A22B leads
Llama 2-13B: 32.3 (#269), Qwen3 235B-A22B: 46.1 (#26)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1064 | 1426 |
| Fiction.LiveBench | — | 75% |
Writing & Preference Qwen3 235B-A22B leads
Llama 2-13B: 29.8 (#289), Qwen3 235B-A22B: 59.6 (#108)
| Benchmark | Llama 2-13B | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1084 | 1419 |
| LMArena Creative Writing | 1047 | 1384 |
| LMArena Multi-Turn | 1050 | 1432 |
| Short-Story Creative Writing | — | 83% |
| EQ-Bench Creative Writing | — | 1366 |
| WildBench | — | 86.6% |
Frequently asked questions
Is Llama 2-13B better than Qwen3 235B-A22B?
Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 29.6 on the Noometry Index.
Is Llama 2-13B or Qwen3 235B-A22B better for coding?
Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 30.9 in the Noometry coding category.
How many benchmarks do Llama 2-13B and Qwen3 235B-A22B share?
20 benchmarks have published results for both models. Llama 2-13B has 32 scored results on Noometry and Qwen3 235B-A22B has 49.