Model comparison
Llama 3-8B vs Qwen3-30B-A3B
Qwen3-30B-A3B is the stronger model overall, scoring 38.5 to 25.3 on the Noometry Index. Llama 3-8B costs 1.5× less per token, which makes it the better buy when Qwen3-30B-A3B's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Llama 3-8B scores higher in 1 category and Qwen3-30B-A3B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3-30B-A3B leads 41.7 to 7.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.9% for Llama 3-8B and 70.3% for Qwen3-30B-A3B.
- Llama 3-8B is cheaper at $0.14 / $0.14 per million input/output tokens, against $0.12 / $0.50 for Qwen3-30B-A3B.
- Qwen3-30B-A3B accepts more context: 41K tokens versus 8K.
Side by side
| Llama 3-8B | Qwen3-30B-A3B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 25.3 | 38.5 |
| Released | 2024-04-18 | 2025-04-28 |
| Weights | Open | Open |
| Context window | 8K | 41K |
| Max output | 8K | 16K |
| Input $ / M tokens | $0.14 | $0.12 |
| Output $ / M tokens | $0.14 | $0.50 |
| Results tracked | 34 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3-30B-A3B leads
Llama 3-8B: 30.5 (#285), Qwen3-30B-A3B: 37.0 (#188)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Coding | 1152 | 1416 |
| SciCode | — | 33.3% |
| WeirdML | — | 29.8% |
| BigCodeBench Instruct | 31.9% | — |
| BigCodeBench Complete | 36.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Agentic & Tool Use Not comparable
Llama 3-8B: —, Qwen3-30B-A3B: 29.5 (#75)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41.4% |
Reasoning Qwen3-30B-A3B leads
Llama 3-8B: 13.8 (#321), Qwen3-30B-A3B: 21.6 (#193)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| Chess Puzzles | 0% | 8% |
| LMArena Hard Prompts | 1133 | 1398 |
| DTBench | 43.9% | 69.3% |
| Epoch Capabilities Index | 116.33 | 139.63 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| LMCA | — | 22.4% |
| Adversarial NLI | 57.3% | — |
| ForecastBench | 58.6 | — |
| WinoGrande | 75.7% | — |
Math Qwen3-30B-A3B leads
Llama 3-8B: 8.3 (#317), Qwen3-30B-A3B: 36.0 (#133)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | 70.3% |
| LMArena Math | 1151 | 1394 |
| MathArena Final-Answer Competitions | — | 47.8% |
| MATH Level 5 | 6.1% | — |
Knowledge Qwen3-30B-A3B leads
Llama 3-8B: 7.8 (#303), Qwen3-30B-A3B: 41.7 (#96)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 26.1% | 70.1% |
| LMArena Expert | 1113 | 1395 |
| Confabulations | — | 12.3% |
| ARC (AI2) Challenge | 82.8% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |
Multilingual Qwen3-30B-A3B leads
Llama 3-8B: 31.0 (#256), Qwen3-30B-A3B: 49.8 (#126)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1098 | 1372 |
| LMArena Chinese | 1076 | 1434 |
| LMArena French | 1158 | 1419 |
| LMArena German | 1104 | 1379 |
| LMArena Japanese | 967 | 1338 |
| LMArena Korean | 1003 | 1333 |
| LMArena Russian | 1109 | 1369 |
| LMArena Spanish | 1172 | 1403 |
Instruction Following Qwen3-30B-A3B leads
Llama 3-8B: 58.7 (#255), Qwen3-30B-A3B: 72.2 (#137)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1127 | 1364 |
Long Context Llama 3-8B leads
Llama 3-8B: 34.2 (#246), Qwen3-30B-A3B: 31.0 (#278)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Longer Query | 1128 | 1379 |
| Fiction.LiveBench | — | 40.6% |
Writing & Preference Qwen3-30B-A3B leads
Llama 3-8B: 37.7 (#251), Qwen3-30B-A3B: 55.9 (#138)
| Benchmark | Llama 3-8B | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1166 | 1384 |
| LMArena Creative Writing | 1150 | 1318 |
| LMArena Multi-Turn | 1152 | 1378 |
| Short-Story Creative Writing | — | 75.3% |
Frequently asked questions
Is Llama 3-8B better than Qwen3-30B-A3B?
Qwen3-30B-A3B is the stronger model overall, scoring 38.5 to 25.3 on the Noometry Index. Llama 3-8B costs 1.5× less per token, which makes it the better buy when Qwen3-30B-A3B's lead doesn't matter for your workload.
Which is cheaper, Llama 3-8B or Qwen3-30B-A3B?
Llama 3-8B is cheaper. It lists at $0.14 per million input tokens and $0.14 per million output tokens; Qwen3-30B-A3B lists at $0.12 and $0.50.
Is Llama 3-8B or Qwen3-30B-A3B better for coding?
Qwen3-30B-A3B scores higher on coding benchmarks: 37.0 versus 30.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3-30B-A3B does, with 41K tokens against 8K.
How many benchmarks do Llama 3-8B and Qwen3-30B-A3B share?
22 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Qwen3-30B-A3B has 32.