All comparisons
Google-Proof Q&A 65.4 — HellaSwag 95.6 — HumanEval 92.7 — Massive Multitask Language Understanding 88.2 —
Qwen2.5 72B Instruct vs Llama 3.3 70B
Side-by-side comparison of specifications, capabilities, and pricing.
| Released | 2024-06-07 | 2025-12-09 |
| Context window | 128K | 8K |
| Parameters | 72.7B | 70B |
| Architecture | decoder only | decoder only |
| License | Unknown | True |
| Knowledge cutoff | — | 2024-12 |
Capabilities | ||
| Multimodal | ||
| Function calling | ||
| Tool use | ||
| JSON mode | ||
Availability | ||
| Providers | — | |
Benchmarks