Model comparison
Llama 3.1-8B vs Qwen2.5 7B Instruct
Qwen2.5 7B Instruct is the stronger model overall, scoring 29.0 to 23.0 on the Noometry Index. Llama 3.1-8B costs 5.3× less per token, which makes it the better buy when Qwen2.5 7B Instruct's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Llama 3.1-8B scores higher in 1 category and Qwen2.5 7B Instruct in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5 7B Instruct leads 48.8 to 29.7.
- The biggest single-benchmark swing is Omni-MATH: 13.7% for Llama 3.1-8B and 29.4% for Qwen2.5 7B Instruct.
- Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $0.17 / $0.70 for Qwen2.5 7B Instruct.
- Qwen2.5 7B Instruct accepts more context: 131K tokens versus 128K.
Side by side
| Llama 3.1-8B | Qwen2.5 7B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 23.0 | 29.0 |
| Released | 2024-07-23 | 2024-09 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 4K | 8K |
| Input $ / M tokens | $0.05 | $0.17 |
| Output $ / M tokens | $0.08 | $0.70 |
| Results tracked | 43 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 7B Instruct leads
Llama 3.1-8B: 20.2 (#340), Qwen2.5 7B Instruct: 36.5 (#208)
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| BigCodeBench Instruct | 32.8% | 37.6% |
| BigCodeBench Complete | 40.5% | 46.1% |
| SciCode | 13.2% | — |
| WeirdML | 1.7% | — |
| LMArena Coding | 1195 | — |
| HumanEval+ | 62.8% | — |
| MBPP+ | 55.6% | — |
Agentic & Tool Use Qwen2.5 7B Instruct leads
Llama 3.1-8B: 22.5 (#131), Qwen2.5 7B Instruct: 23.8 (#124)
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| BALROG | 15.1% | 7.8% |
| Berkeley Function Calling Leaderboard | 25.8% | — |
Reasoning Too close to call
Llama 3.1-8B: 14.9 (#321), Qwen2.5 7B Instruct: 14.8 (#322)
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| DTBench | 50.9% | 47.7% |
| LMCA | 5.4% | 6.4% |
| Epoch Capabilities Index | 116.57 | 118.51 |
| CritPt | 0% | — |
| LMArena Hard Prompts | 1175 | — |
| PIQA | 81.2% | — |
Math Qwen2.5 7B Instruct leads
Llama 3.1-8B: 10.2 (#317), Qwen2.5 7B Instruct: 12.6 (#306)
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.7% | 2.5% |
| Omni-MATH | 13.7% | 29.4% |
| LMArena Math | 1179 | — |
| MATH Level 5 | 22.9% | — |
| GSM8K | 82.4% | — |
Knowledge Qwen2.5 7B Instruct leads
Llama 3.1-8B: 8.0 (#307), Qwen2.5 7B Instruct: 17.0 (#286)
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 27% | 35.5% |
| MMLU-Pro | 40.6% | 53.9% |
| GPQA (HELM) | 24.7% | 34.1% |
| MMLU | 56.1% | 72.9% |
| LMArena Expert | 1144 | — |
| BoolQ | 82.8% | — |
Multilingual Not comparable
Llama 3.1-8B: 34.0 (#249), Qwen2.5 7B Instruct: —
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Non-English | 1148 | — |
| LMArena Chinese | 1151 | — |
| LMArena French | 1177 | — |
| LMArena German | 1144 | — |
| LMArena Japanese | 1061 | — |
| LMArena Korean | 1053 | — |
| LMArena Russian | 1158 | — |
| LMArena Spanish | 1169 | — |
Instruction Following Qwen2.5 7B Instruct leads
Llama 3.1-8B: 58.9 (#258), Qwen2.5 7B Instruct: 63.2 (#231)
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| IFEval | 74.3% | 74.1% |
| LMArena Instruction Following | 1159 | — |
Long Context Not comparable
Llama 3.1-8B: 35.8 (#238), Qwen2.5 7B Instruct: —
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Longer Query | 1182 | — |
Writing & Preference Qwen2.5 7B Instruct leads
Llama 3.1-8B: 29.7 (#290), Qwen2.5 7B Instruct: 48.8 (#195)
| Benchmark | Llama 3.1-8B | Qwen2.5 7B Instruct |
|---|---|---|
| WildBench | 68.7% | 73.1% |
| LMArena Text | 1187 | — |
| LMArena Creative Writing | 1154 | — |
| EQ-Bench Creative Writing | 713 | — |
| LMArena Multi-Turn | 1172 | — |
Frequently asked questions
Is Llama 3.1-8B better than Qwen2.5 7B Instruct?
Qwen2.5 7B Instruct is the stronger model overall, scoring 29.0 to 23.0 on the Noometry Index. Llama 3.1-8B costs 5.3× less per token, which makes it the better buy when Qwen2.5 7B Instruct's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-8B or Qwen2.5 7B Instruct?
Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Qwen2.5 7B Instruct lists at $0.17 and $0.70.
Is Llama 3.1-8B or Qwen2.5 7B Instruct better for coding?
Qwen2.5 7B Instruct scores higher on coding benchmarks: 36.5 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Qwen2.5 7B Instruct does, with 131K tokens against 128K.
How many benchmarks do Llama 3.1-8B and Qwen2.5 7B Instruct share?
15 benchmarks have published results for both models. Llama 3.1-8B has 43 scored results on Noometry and Qwen2.5 7B Instruct has 15.