Model comparison
Grok 4 vs Llama2 70b Steerlm Chat
Grok 4 is the stronger model overall, scoring 48.1 to 31.8 on the Noometry Index.
Last verified . 9 shared benchmarks.
NVIDIASummary
- They share 9 benchmarks with published results for both. Grok 4 scores higher in 7 categories and Llama2 70b Steerlm Chat in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 30.4.
- Llama2 70b Steerlm Chat has downloadable open weights; the other is API-only.
Side by side
| Grok 4 | Llama2 70b Steerlm Chat | |
|---|---|---|
| Provider | xAI | NVIDIA |
| Noometry Index | 48.1 | 31.8 |
| Released | 2025-07-09 | — |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 48 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4 leads
Grok 4: 50.3 (#46), Llama2 70b Steerlm Chat: 29.9 (#300)
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Coding | 1408 | 1025 |
| Aider Polyglot | 79.6% | — |
| WeirdML | 45.7% | — |
Agentic & Tool Use Not comparable
Grok 4: 32.3 (#68), Llama2 70b Steerlm Chat: —
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| Terminal-Bench | 27.2% | — |
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 21.1% | — |
| Cybench | 43% | — |
| DeepResearch Bench | 47.3% | — |
| BALROG | 43.6% | — |
| LMArena Search | 1142 | — |
| METR Time Horizons | 66.6% | — |
Reasoning Grok 4 leads
Grok 4: 36.7 (#65), Llama2 70b Steerlm Chat: 20.0 (#246)
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Hard Prompts | 1409 | 1047 |
| ARC-AGI-2 | 16% | — |
| SimpleBench | 60.5% | — |
| Kagi LLM Benchmark | 73.6% | — |
| ARC-AGI-1 | 66.7% | — |
| Chess Puzzles | 28% | — |
| Epoch Capabilities Index | 146.44 | — |
| ForecastBench | 60.9 | — |
Math Grok 4 leads
Grok 4: 48.4 (#64), Llama2 70b Steerlm Chat: 31.3 (#226)
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Math | 1422 | 1072 |
| OTIS Mock AIME 2024-2025 | 84% | — |
| Omni-MATH | 60.3% | — |
| FrontierMath (Feb 2025 set) | 19.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Not comparable
Grok 4: 53.8 (#55), Llama2 70b Steerlm Chat: —
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| GPQA Diamond | 87% | — |
| MMLU-Pro | 85.1% | — |
| Confabulations | 12.4% | — |
| GPQA (HELM) | 72.7% | — |
| LMArena Expert | 1415 | — |
Multimodal Not comparable
Grok 4: 33.7 (#94), Llama2 70b Steerlm Chat: —
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Vision | 1210 | — |
| GeoBench | 45% | — |
Multilingual Grok 4 leads
Grok 4: 51.8 (#103), Llama2 70b Steerlm Chat: 28.8 (#270)
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Non-English | 1403 | 1063 |
| LMArena Chinese | 1427 | — |
| LMArena French | 1418 | — |
| LMArena German | 1429 | — |
| LMArena Japanese | 1394 | — |
| LMArena Korean | 1377 | — |
| LMArena Russian | 1410 | — |
| LMArena Spanish | 1420 | — |
Instruction Following Grok 4 leads
Grok 4: 79.2 (#5), Llama2 70b Steerlm Chat: 54.2 (#279)
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Instruction Following | 1387 | 1060 |
| IFEval | 94.9% | — |
Long Context Grok 4 leads
Grok 4: 63.1 (#4), Llama2 70b Steerlm Chat: 30.4 (#288)
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Longer Query | 1409 | 998 |
| Fiction.LiveBench | 94.4% | — |
Writing & Preference Grok 4 leads
Grok 4: 58.5 (#116), Llama2 70b Steerlm Chat: 31.6 (#283)
| Benchmark | Grok 4 | Llama2 70b Steerlm Chat |
|---|---|---|
| LMArena Text | 1411 | 1098 |
| LMArena Creative Writing | 1397 | 1091 |
| LMArena Multi-Turn | 1416 | 1058 |
| Short-Story Creative Writing | 76.9% | — |
| WildBench | 79.7% | — |
Frequently asked questions
Is Grok 4 better than Llama2 70b Steerlm Chat?
Grok 4 is the stronger model overall, scoring 48.1 to 31.8 on the Noometry Index.
Is Grok 4 or Llama2 70b Steerlm Chat better for coding?
Grok 4 scores higher on coding benchmarks: 50.3 versus 29.9 in the Noometry coding category.
How many benchmarks do Grok 4 and Llama2 70b Steerlm Chat share?
9 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Llama2 70b Steerlm Chat has 9.