Grok 4.20 0309 vs Llama 3.1 Nemotron 70B Instruct
xAI vs NVIDIA — benchmarks, pricing, and capabilities side by side.
- •Grok 4.20 0309 has the higher intelligence index (88.5 vs 43.7)
- •Llama 3.1 Nemotron 70B Instruct is cheaper ($1.20 vs $2.00 per 1M input)
- •Llama 3.1 Nemotron 70B Instruct is faster
| Grok 4.20 0309 | Llama 3.1 Nemotron 70B Instruct | |
|---|---|---|
| Intelligence index | 88.5 | 43.7 |
| Developer | xAI | NVIDIA |
| Type | LLM | LLM |
| Access | — | Open weights |
| Context window | — | — |
| Input price | $2.00 / 1M | $1.20 / 1M |
| Output price | $6.00 / 1M | $1.20 / 1M |
| Speed | 97 tok/s | 292 tok/s |
| Released | March 10, 2026 | October 1, 2024 |
| Parameters | — | 70000000000 |
| Input modalities | — | — |
| Output modalities | — | — |
Shared benchmarks
Grok 4.20 0309
Llama 3.1 Nemotron 70B Instruct
GPQA Diamond
88.5
46.5
Humanity’s Last Exam
30
4.6