Model comparison
Grok 3 vs Llama2 70b Steerlm Chat
Grok 3 is the stronger model overall, scoring 39.9 to 31.8 on the Noometry Index.
Last verified . 9 shared benchmarks.
NVIDIASummary
- They share 9 benchmarks with published results for both. Grok 3 scores higher in 6 categories and Llama2 70b Steerlm Chat in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Grok 3 leads 55.8 to 31.6.
- Llama2 70b Steerlm Chat has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | Llama2 70b Steerlm Chat | |
|---|---|---|
| Provider | xAI | NVIDIA |
| Noometry Index | 39.9 | 31.8 |
| Released | 2025-04-09 | — |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 40 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Grok 3: 41.9 (#115), Llama2 70b Steerlm Chat: 29.9 (#300)
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Coding | 1432 | 1025 |
| Aider Polyglot | 53.3% | — |
| WeirdML | 37.2% | — |
Agentic & Tool Use Not comparable
Grok 3: 30.5 (#76), Llama2 70b Steerlm Chat: —
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| BALROG | 29.5% | — |
Reasoning Llama2 70b Steerlm Chat leads
Grok 3: 13.7 (#333), Llama2 70b Steerlm Chat: 20.0 (#246)
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Hard Prompts | 1434 | 1047 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| Kagi LLM Benchmark | 61.3% | — |
| ARC-AGI-1 | 5.5% | — |
| Epoch Capabilities Index | 138.33 | — |
Math Grok 3 leads
Grok 3: 38.0 (#145), Llama2 70b Steerlm Chat: 31.3 (#226)
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Math | 1391 | 1072 |
| OTIS Mock AIME 2024-2025 | 55.6% | — |
| Omni-MATH | 46.4% | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Not comparable
Grok 3: 46.2 (#82), Llama2 70b Steerlm Chat: —
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |
| LMArena Expert | 1421 | — |
Multilingual Grok 3 leads
Grok 3: 52.3 (#87), Llama2 70b Steerlm Chat: 28.8 (#270)
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Non-English | 1410 | 1063 |
| LMArena Chinese | 1448 | — |
| LMArena French | 1460 | — |
| LMArena German | 1431 | — |
| LMArena Japanese | 1387 | — |
| LMArena Korean | 1373 | — |
| LMArena Russian | 1416 | — |
| LMArena Spanish | 1417 | — |
Instruction Following Grok 3 leads
Grok 3: 75.0 (#73), Llama2 70b Steerlm Chat: 54.2 (#279)
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Instruction Following | 1409 | 1060 |
| IFEval | 88.4% | — |
Long Context Grok 3 leads
Grok 3: 38.7 (#192), Llama2 70b Steerlm Chat: 30.4 (#288)
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Longer Query | 1439 | 998 |
| Fiction.LiveBench | 58.3% | — |
Writing & Preference Grok 3 leads
Grok 3: 55.8 (#141), Llama2 70b Steerlm Chat: 31.6 (#283)
| Benchmark | Grok 3 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Text | 1426 | 1098 |
| LMArena Creative Writing | 1414 | 1091 |
| LMArena Multi-Turn | 1425 | 1058 |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
Frequently asked questions
Is Grok 3 better than Llama2 70b Steerlm Chat?
Grok 3 is the stronger model overall, scoring 39.9 to 31.8 on the Noometry Index.
Is Grok 3 or Llama2 70b Steerlm Chat better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 29.9 in the Noometry coding category.
How many benchmarks do Grok 3 and Llama2 70b Steerlm Chat share?
9 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Llama2 70b Steerlm Chat has 9.