Model comparison
Llama 2-70B vs Yi-34B
Yi-34B is the stronger model overall, scoring 27.8 to 24.4 on the Noometry Index.
Last verified . 23 shared benchmarks.
MetaSummary
- They share 23 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Yi-34B in 8 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Yi-34B leads 21.6 to 8.1.
- The biggest single-benchmark swing is GPQA Diamond: 26.3% for Llama 2-70B and 14.7% for Yi-34B.
Side by side
| Llama 2-70B | Yi-34B | |
|---|---|---|
| Provider | Meta | 01.AI |
| Noometry Index | 24.4 | 27.8 |
| Released | 2023-07-18 | 2023-11-02 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Llama 2-70B: 31.4 (#286), Yi-34B: 32.3 (#274)
| Benchmark | Llama 2-70B | Yi-34B |
|---|---|---|
| LMArena Coding | 1079 | 1112 |
Reasoning Yi-34B leads
Llama 2-70B: 14.4 (#325), Yi-34B: 21.2 (#226)
| Benchmark | Llama 2-70B | Yi-34B |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1104 |
| BIG-Bench Hard | 64.9% | 71.7% |
| Epoch Capabilities Index | 113.79 | 117.39 |
| DTBench | 41.6% | — |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math Yi-34B leads
Llama 2-70B: 8.1 (#326), Yi-34B: 21.6 (#282)
| Benchmark | Llama 2-70B | Yi-34B |
|---|---|---|
| LMArena Math | 1091 | 1114 |
| MATH Level 5 | 3.3% | 5.1% |
| GSM8K | 69.6% | 76% |
| OTIS Mock AIME 2024-2025 | 0% | — |
Knowledge Too close to call
Llama 2-70B: 7.4 (#310), Yi-34B: 7.5 (#309)
| Benchmark | Llama 2-70B | Yi-34B |
|---|---|---|
| GPQA Diamond | 26.3% | 14.7% |
| LMArena Expert | 1039 | 1061 |
| MMLU | 69.9% | 76.3% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Yi-34B leads
Llama 2-70B: 27.7 (#274), Yi-34B: 29.7 (#264)
| Benchmark | Llama 2-70B | Yi-34B |
|---|---|---|
| LMArena Non-English | 1045 | 1079 |
| LMArena Chinese | 995 | 1176 |
| LMArena French | 1090 | 1081 |
| LMArena German | 1041 | 1042 |
| LMArena Japanese | 927 | 993 |
| LMArena Korean | 964 | 959 |
| LMArena Russian | 1083 | 1050 |
| LMArena Spanish | 1143 | 1070 |
Instruction Following Yi-34B leads
Llama 2-70B: 54.9 (#278), Yi-34B: 56.2 (#274)
| Benchmark | Llama 2-70B | Yi-34B |
|---|---|---|
| LMArena Instruction Following | 1071 | 1091 |
Long Context Too close to call
Llama 2-70B: 32.3 (#270), Yi-34B: 33.2 (#264)
| Benchmark | Llama 2-70B | Yi-34B |
|---|---|---|
| LMArena Longer Query | 1062 | 1094 |
Writing & Preference Yi-34B leads
Llama 2-70B: 32.3 (#279), Yi-34B: 34.1 (#273)
| Benchmark | Llama 2-70B | Yi-34B |
|---|---|---|
| LMArena Text | 1115 | 1129 |
| LMArena Creative Writing | 1075 | 1108 |
| LMArena Multi-Turn | 1088 | 1113 |
Frequently asked questions
Is Llama 2-70B better than Yi-34B?
Yi-34B is the stronger model overall, scoring 27.8 to 24.4 on the Noometry Index.
Is Llama 2-70B or Yi-34B better for coding?
They score almost the same on coding (31.4 vs 32.3); test both on your own repository before choosing.
How many benchmarks do Llama 2-70B and Yi-34B share?
23 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Yi-34B has 23.