Model comparison
Llama 3-8B vs Qwen3.5-Flash
Qwen3.5-Flash is the stronger model overall, scoring 42.5 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-Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3.5-Flash leads 43.2 to 7.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.9% for Llama 3-8B and 84.4% for Qwen3.5-Flash.
- Llama 3-8B has downloadable open weights; the other is API-only.
Side by side
| Llama 3-8B | Qwen3.5-Flash | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 25.5 | 42.5 |
| Released | 2024-04-18 | 2026-02-23 |
| Weights | Open | Proprietary |
| Context window | — | 1M |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.40 |
| Results tracked | 34 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5-Flash leads
Llama 3-8B: 31.0 (#289), Qwen3.5-Flash: 34.2 (#242)
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| LMArena Coding | 1152 | 1412 |
| LMArena WebDev | — | 1244 |
| BigCodeBench Instruct | 31.9% | — |
| BigCodeBench Complete | 36.9% | — |
| ALE-Bench | — | 221.8 |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Agentic & Tool Use Not comparable
Llama 3-8B: —, Qwen3.5-Flash: —
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| Vending-Bench 2 | — | 462.69 |
Reasoning Qwen3.5-Flash leads
Llama 3-8B: 14.3 (#326), Qwen3.5-Flash: 33.7 (#72)
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| Chess Puzzles | 0% | 21% |
| LMArena Hard Prompts | 1133 | 1403 |
| DTBench | 43.9% | 82.9% |
| Epoch Capabilities Index | 116.45 | 143.98 |
| Mystery Game Puzzles | — | 20% |
| LMCA | — | 29.1% |
| Adversarial NLI | 57.3% | — |
| ForecastBench | 58.6 | — |
| WinoGrande | 75.7% | — |
Math Qwen3.5-Flash leads
Llama 3-8B: 8.8 (#323), Qwen3.5-Flash: 37.4 (#158)
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | 84.4% |
| LMArena Math | 1151 | 1407 |
| FrontierMath (Tiers 1-3) | — | 18.2% |
| MATH Level 5 | 6.1% | — |
| FrontierMath (Feb 2025 set) | — | 6.2% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Qwen3.5-Flash leads
Llama 3-8B: 7.8 (#308), Qwen3.5-Flash: 43.2 (#93)
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| GPQA Diamond | 26.1% | 82.3% |
| LMArena Expert | 1113 | 1407 |
| SimpleQA Verified | — | 20.3% |
| Vectara Hallucination Rate | — | 10.5% |
| ARC (AI2) Challenge | 82.8% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |
Multilingual Qwen3.5-Flash leads
Llama 3-8B: 30.8 (#261), Qwen3.5-Flash: 50.5 (#121)
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| LMArena Non-English | 1098 | 1385 |
| LMArena Chinese | 1076 | 1446 |
| LMArena French | 1159 | 1412 |
| LMArena German | 1104 | 1390 |
| LMArena Japanese | 967 | 1368 |
| LMArena Korean | 1004 | 1344 |
| LMArena Russian | 1109 | 1379 |
| LMArena Spanish | 1173 | 1400 |
Instruction Following Qwen3.5-Flash leads
Llama 3-8B: 58.4 (#260), Qwen3.5-Flash: 72.6 (#139)
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| LMArena Instruction Following | 1127 | 1374 |
Long Context Qwen3.5-Flash leads
Llama 3-8B: 34.2 (#251), Qwen3.5-Flash: 42.4 (#124)
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| LMArena Longer Query | 1128 | 1392 |
Writing & Preference Qwen3.5-Flash leads
Llama 3-8B: 37.5 (#256), Qwen3.5-Flash: 57.9 (#122)
| Benchmark | Llama 3-8B | Qwen3.5-Flash |
|---|---|---|
| LMArena Text | 1166 | 1397 |
| LMArena Creative Writing | 1150 | 1343 |
| LMArena Multi-Turn | 1152 | 1393 |
Frequently asked questions
Is Llama 3-8B better than Qwen3.5-Flash?
Qwen3.5-Flash is the stronger model overall, scoring 42.5 to 25.5 on the Noometry Index.
Is Llama 3-8B or Qwen3.5-Flash better for coding?
Qwen3.5-Flash scores higher on coding benchmarks: 34.2 versus 31.0 in the Noometry coding category.
How many benchmarks do Llama 3-8B and Qwen3.5-Flash share?
22 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Qwen3.5-Flash has 32.