Model comparison
Grok-3 mini vs Llama2 70b Steerlm Chat
Grok-3 mini is the stronger model overall, scoring 41.0 to 31.2 on the Noometry Index.
Last verified . 9 shared benchmarks.
NVIDIASummary
- They share 9 benchmarks with published results for both. Grok-3 mini 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 instruction following, where Grok-3 mini leads 78.7 to 54.5.
- Llama2 70b Steerlm Chat has downloadable open weights; the other is API-only.
Side by side
| Grok-3 mini | Llama2 70b Steerlm Chat | |
|---|---|---|
| Provider | xAI | NVIDIA |
| Noometry Index | 41.0 | 31.2 |
| Released | 2025-04-09 | — |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok-3 mini leads
Grok-3 mini: 40.4 (#126), Llama2 70b Steerlm Chat: 29.3 (#295)
| Benchmark | Grok-3 mini | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Coding | 1380 | 1025 |
| Aider Polyglot | 49.3% | — |
| WeirdML | 42.6% | — |
Reasoning Llama2 70b Steerlm Chat leads
Grok-3 mini: 13.8 (#323), Llama2 70b Steerlm Chat: 19.1 (#245)
| Benchmark | Grok-3 mini | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Hard Prompts | 1374 | 1047 |
| ARC-AGI-2 | 0.4% | — |
| Kagi LLM Benchmark | 61.3% | — |
| ARC-AGI-1 | 16.5% | — |
| Epoch Capabilities Index | 140.35 | — |
Math Grok-3 mini leads
Grok-3 mini: 40.3 (#78), Llama2 70b Steerlm Chat: 28.8 (#223)
| Benchmark | Grok-3 mini | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Math | 1386 | 1072 |
| OTIS Mock AIME 2024-2025 | 77.8% | — |
| Omni-MATH | 31.8% | — |
| MATH Level 5 | 90.9% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
Knowledge Not comparable
Grok-3 mini: 46.3 (#75), Llama2 70b Steerlm Chat: —
| Benchmark | Grok-3 mini | Llama2 70b Steerlm Chat |
|---|---|---|
| GPQA Diamond | 76.3% | — |
| MMLU-Pro | 79.9% | — |
| Confabulations | 10.8% | — |
| GPQA (HELM) | 67.5% | — |
| LMArena Expert | 1395 | — |
Multilingual Grok-3 mini leads
Grok-3 mini: 48.3 (#141), Llama2 70b Steerlm Chat: 28.9 (#265)
| Benchmark | Grok-3 mini | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Non-English | 1352 | 1064 |
| LMArena Chinese | 1387 | — |
| LMArena French | 1357 | — |
| LMArena German | 1350 | — |
| LMArena Japanese | 1343 | — |
| LMArena Korean | 1336 | — |
| LMArena Russian | 1352 | — |
| LMArena Spanish | 1379 | — |
Instruction Following Grok-3 mini leads
Grok-3 mini: 78.7 (#8), Llama2 70b Steerlm Chat: 54.5 (#274)
| Benchmark | Grok-3 mini | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Instruction Following | 1356 | 1061 |
| IFEval | 95.1% | — |
Long Context Grok-3 mini leads
Grok-3 mini: 41.0 (#142), Llama2 70b Steerlm Chat: 30.4 (#283)
| Benchmark | Grok-3 mini | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Longer Query | 1372 | 998 |
| Fiction.LiveBench | 66.7% | — |
Writing & Preference Grok-3 mini leads
Grok-3 mini: 52.7 (#163), Llama2 70b Steerlm Chat: 31.8 (#278)
| Benchmark | Grok-3 mini | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Text | 1370 | 1098 |
| LMArena Creative Writing | 1342 | 1091 |
| LMArena Multi-Turn | 1355 | 1058 |
| Short-Story Creative Writing | 73.5% | — |
| WildBench | 65.1% | — |
Frequently asked questions
Is Grok-3 mini better than Llama2 70b Steerlm Chat?
Grok-3 mini is the stronger model overall, scoring 41.0 to 31.2 on the Noometry Index.
Is Grok-3 mini or Llama2 70b Steerlm Chat better for coding?
Grok-3 mini scores higher on coding benchmarks: 40.4 versus 29.3 in the Noometry coding category.
How many benchmarks do Grok-3 mini and Llama2 70b Steerlm Chat share?
9 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Llama2 70b Steerlm Chat has 9.