Model comparison
Llama 3-70B vs Mercury 2
Mercury 2 is the stronger model overall, scoring 39.1 to 28.8 on the Noometry Index.
Last verified . 11 shared benchmarks.
MetaSummary
- They share 11 benchmarks with published results for both. Llama 3-70B scores higher in 1 category and Mercury 2 in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mercury 2 leads 36.2 to 20.8.
- Llama 3-70B has downloadable open weights; the other is API-only.
Side by side
| Llama 3-70B | Mercury 2 | |
|---|---|---|
| Provider | Meta | Inception |
| Noometry Index | 28.8 | 39.1 |
| Released | 2024-04-18 | 2026-02-20 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 50K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $0.75 |
| Results tracked | 31 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-70B leads
Llama 3-70B: 35.8 (#218), Mercury 2: 33.5 (#255)
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| LMArena Coding | 1206 | 1391 |
| LMArena WebDev | — | 1171 |
| SciCode | — | 38.7% |
| WeirdML | — | 43.2% |
| BigCodeBench Instruct | 43.6% | — |
| BigCodeBench Complete | 54.5% | — |
| ALE-Bench | — | 785.58 |
| HumanEval+ | 72% | — |
| MBPP+ | 69% | — |
Agentic & Tool Use Not comparable
Llama 3-70B: 21.1 (#139), Mercury 2: —
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| Cybench | 5% | — |
Reasoning Mercury 2 leads
Llama 3-70B: 18.0 (#288), Mercury 2: 23.8 (#170)
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| LMArena Hard Prompts | 1195 | 1362 |
| Kagi LLM Benchmark | 35.1% | — |
| CritPt | — | 0.8% |
| DTBench | 54.2% | — |
| Epoch Capabilities Index | 122.93 | — |
| ForecastBench | 57.1 | — |
| WinoGrande | 83.5% | — |
Math Not comparable
Llama 3-70B: 12.8 (#305), Mercury 2: —
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | — |
| LMArena Math | 1218 | — |
| MATH Level 5 | 22.6% | — |
Knowledge Mercury 2 leads
Llama 3-70B: 20.8 (#277), Mercury 2: 36.2 (#172)
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| LMArena Expert | 1149 | 1358 |
| GPQA Diamond | 40.6% | — |
| Vectara Hallucination Rate | — | 12.3% |
| MMLU | 79.3% | — |
Multilingual Mercury 2 leads
Llama 3-70B: 33.6 (#251), Mercury 2: 46.6 (#157)
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| LMArena Non-English | 1142 | 1331 |
| LMArena Chinese | 1114 | 1417 |
| LMArena Russian | 1159 | 1304 |
| LMArena French | 1232 | — |
| LMArena German | 1169 | — |
| LMArena Japanese | 1017 | — |
| LMArena Korean | 1017 | — |
| LMArena Spanish | 1241 | — |
Instruction Following Mercury 2 leads
Llama 3-70B: 62.5 (#238), Mercury 2: 70.2 (#165)
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| LMArena Instruction Following | 1194 | 1329 |
Long Context Mercury 2 leads
Llama 3-70B: 35.6 (#240), Mercury 2: 40.5 (#154)
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| LMArena Longer Query | 1174 | 1330 |
Writing & Preference Mercury 2 leads
Llama 3-70B: 42.8 (#231), Mercury 2: 53.8 (#155)
| Benchmark | Llama 3-70B | Mercury 2 |
|---|---|---|
| LMArena Text | 1221 | 1355 |
| LMArena Creative Writing | 1210 | 1289 |
| LMArena Multi-Turn | 1223 | 1358 |
Frequently asked questions
Is Llama 3-70B better than Mercury 2?
Mercury 2 is the stronger model overall, scoring 39.1 to 28.8 on the Noometry Index.
Is Llama 3-70B or Mercury 2 better for coding?
Llama 3-70B scores higher on coding benchmarks: 35.8 versus 33.5 in the Noometry coding category.
How many benchmarks do Llama 3-70B and Mercury 2 share?
11 benchmarks have published results for both models. Llama 3-70B has 31 scored results on Noometry and Mercury 2 has 17.