Model comparison
Llama 3-8B vs Qwen3.5 35B-A3B
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 25.5 on the Noometry Index.
Last verified . 22 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Llama 3-8B scores higher in 0 categories and Qwen3.5 35B-A3B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5 35B-A3B leads 47.8 to 7.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.9% for Llama 3-8B and 70% for Qwen3.5 35B-A3B.
Side by side
| Llama 3-8B | Qwen3.5 35B-A3B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 25.5 | 42.0 |
| Released | 2024-04-18 | 2026-02-01 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $2 |
| Results tracked | 34 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 35B-A3B leads
Llama 3-8B: 31.0 (#289), Qwen3.5 35B-A3B: 33.8 (#251)
| Benchmark | Llama 3-8B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Coding | 1152 | 1410 |
| LMArena WebDev | — | 1254 |
| SciCode | — | 29.3% |
| BigCodeBench Instruct | 31.9% | — |
| BigCodeBench Complete | 36.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Reasoning Qwen3.5 35B-A3B leads
Llama 3-8B: 14.3 (#326), Qwen3.5 35B-A3B: 24.6 (#161)
| Benchmark | Llama 3-8B | Qwen3.5 35B-A3B |
|---|---|---|
| Chess Puzzles | 0% | 10% |
| LMArena Hard Prompts | 1133 | 1400 |
| DTBench | 43.9% | 80% |
| Epoch Capabilities Index | 116.45 | 142.52 |
| CritPt | — | 0.6% |
| LMCA | — | 29.5% |
| Adversarial NLI | 57.3% | — |
| ForecastBench | 58.6 | — |
| WinoGrande | 75.7% | — |
Math Qwen3.5 35B-A3B leads
Llama 3-8B: 8.8 (#323), Qwen3.5 35B-A3B: 39.9 (#97)
| Benchmark | Llama 3-8B | Qwen3.5 35B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | 70% |
| LMArena Math | 1151 | 1404 |
| MathArena Final-Answer Competitions | — | 56% |
| MATH Level 5 | 6.1% | — |
Knowledge Qwen3.5 35B-A3B leads
Llama 3-8B: 7.8 (#308), Qwen3.5 35B-A3B: 47.8 (#79)
| Benchmark | Llama 3-8B | Qwen3.5 35B-A3B |
|---|---|---|
| GPQA Diamond | 26.1% | 83.5% |
| LMArena Expert | 1113 | 1408 |
| Vectara Hallucination Rate | — | 10.5% |
| ARC (AI2) Challenge | 82.8% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |
Multilingual Qwen3.5 35B-A3B leads
Llama 3-8B: 30.8 (#261), Qwen3.5 35B-A3B: 50.0 (#127)
| Benchmark | Llama 3-8B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Non-English | 1098 | 1378 |
| LMArena Chinese | 1076 | 1457 |
| LMArena French | 1159 | 1412 |
| LMArena German | 1104 | 1367 |
| LMArena Japanese | 967 | 1325 |
| LMArena Korean | 1004 | 1356 |
| LMArena Russian | 1109 | 1376 |
| LMArena Spanish | 1173 | 1392 |
Instruction Following Qwen3.5 35B-A3B leads
Llama 3-8B: 58.4 (#260), Qwen3.5 35B-A3B: 72.8 (#128)
| Benchmark | Llama 3-8B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Instruction Following | 1127 | 1379 |
Long Context Qwen3.5 35B-A3B leads
Llama 3-8B: 34.2 (#251), Qwen3.5 35B-A3B: 42.4 (#127)
| Benchmark | Llama 3-8B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Longer Query | 1128 | 1389 |
Writing & Preference Qwen3.5 35B-A3B leads
Llama 3-8B: 37.5 (#256), Qwen3.5 35B-A3B: 57.9 (#124)
| Benchmark | Llama 3-8B | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Text | 1166 | 1395 |
| LMArena Creative Writing | 1150 | 1346 |
| LMArena Multi-Turn | 1152 | 1390 |
Frequently asked questions
Is Llama 3-8B better than Qwen3.5 35B-A3B?
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 25.5 on the Noometry Index.
Is Llama 3-8B or Qwen3.5 35B-A3B better for coding?
Qwen3.5 35B-A3B scores higher on coding benchmarks: 33.8 versus 31.0 in the Noometry coding category.
How many benchmarks do Llama 3-8B and Qwen3.5 35B-A3B share?
22 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Qwen3.5 35B-A3B has 28.