Model comparison
Llama 3.1-70B vs Qwen2.5 7B Instruct
Llama 3.1-70B and Qwen2.5 7B Instruct score almost the same on the Noometry Index (29.6 vs 29.0), so choose on price, context window or the category you care about most.
Last verified . 14 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. Llama 3.1-70B scores higher in 5 categories and Qwen2.5 7B Instruct in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5 7B Instruct leads 48.8 to 35.4.
- The biggest single-benchmark swing is BALROG: 27.9% for Llama 3.1-70B and 7.8% for Qwen2.5 7B Instruct.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $0.40 / $0.40 for Llama 3.1-70B.
- Qwen2.5 7B Instruct accepts more context: 131K tokens versus 128K.
Side by side
| Llama 3.1-70B | Qwen2.5 7B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 29.6 | 29.0 |
| Released | 2024-07-23 | 2024-09 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 4K | 8K |
| Input $ / M tokens | $0.40 | $0.17 |
| Output $ / M tokens | $0.40 | $0.70 |
| Results tracked | 35 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 7B Instruct leads
Llama 3.1-70B: 30.3 (#296), Qwen2.5 7B Instruct: 36.5 (#208)
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| BigCodeBench Instruct | 46.1% | 37.6% |
| BigCodeBench Complete | 54.8% | 46.1% |
| WeirdML | 9% | — |
| LMArena Coding | 1260 | — |
Agentic & Tool Use Llama 3.1-70B leads
Llama 3.1-70B: 25.1 (#112), Qwen2.5 7B Instruct: 23.8 (#124)
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| BALROG | 27.9% | 7.8% |
| TheAgentCompany | 6.9% | — |
Reasoning Llama 3.1-70B leads
Llama 3.1-70B: 21.6 (#220), Qwen2.5 7B Instruct: 14.8 (#322)
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| DTBench | 60% | 47.7% |
| LMCA | 14.8% | 6.4% |
| Epoch Capabilities Index | 125.92 | 118.51 |
| Chess Puzzles | — | 0% |
| LMArena Hard Prompts | 1241 | — |
Math Too close to call
Llama 3.1-70B: 13.5 (#304), Qwen2.5 7B Instruct: 12.6 (#306)
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 3.6% | 2.5% |
| Omni-MATH | 21% | 29.4% |
| LMArena Math | 1252 | — |
| MATH Level 5 | 36.7% | — |
Knowledge Llama 3.1-70B leads
Llama 3.1-70B: 24.2 (#269), Qwen2.5 7B Instruct: 17.0 (#286)
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 44.2% | 35.5% |
| MMLU-Pro | 65.3% | 53.9% |
| GPQA (HELM) | 42.6% | 34.1% |
| MMLU | 80.1% | 72.9% |
| LMArena Expert | 1209 | — |
Multilingual Not comparable
Llama 3.1-70B: 38.8 (#225), Qwen2.5 7B Instruct: —
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Non-English | 1219 | — |
| LMArena Chinese | 1215 | — |
| LMArena French | 1261 | — |
| LMArena German | 1222 | — |
| LMArena Japanese | 1132 | — |
| LMArena Korean | 1140 | — |
| LMArena Russian | 1234 | — |
| LMArena Spanish | 1253 | — |
Instruction Following Llama 3.1-70B leads
Llama 3.1-70B: 65.3 (#223), Qwen2.5 7B Instruct: 63.2 (#231)
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| IFEval | 82.1% | 74.1% |
| LMArena Instruction Following | 1231 | — |
Long Context Not comparable
Llama 3.1-70B: 37.6 (#214), Qwen2.5 7B Instruct: —
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Longer Query | 1241 | — |
Writing & Preference Qwen2.5 7B Instruct leads
Llama 3.1-70B: 35.4 (#267), Qwen2.5 7B Instruct: 48.8 (#195)
| Benchmark | Llama 3.1-70B | Qwen2.5 7B Instruct |
|---|---|---|
| WildBench | 75.8% | 73.1% |
| LMArena Text | 1261 | — |
| LMArena Creative Writing | 1232 | — |
| EQ-Bench Creative Writing | 784 | — |
| LMArena Multi-Turn | 1256 | — |
Frequently asked questions
Is Llama 3.1-70B better than Qwen2.5 7B Instruct?
Llama 3.1-70B and Qwen2.5 7B Instruct score almost the same on the Noometry Index (29.6 vs 29.0), so choose on price, context window or the category you care about most.
Which is cheaper, Llama 3.1-70B or Qwen2.5 7B Instruct?
Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Llama 3.1-70B lists at $0.40 and $0.40.
Is Llama 3.1-70B or Qwen2.5 7B Instruct better for coding?
Qwen2.5 7B Instruct scores higher on coding benchmarks: 36.5 versus 30.3 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 3.1-70B and Qwen2.5 7B Instruct share?
14 benchmarks have published results for both models. Llama 3.1-70B has 35 scored results on Noometry and Qwen2.5 7B Instruct has 15.