Llama 4 Maverick vs Qwen3 30B-A3B
| Spec | Llama 4 Maverick | Qwen3 30B-A3B |
|---|---|---|
| Respan Index | 14.3 (#152) | 14.8 (#151) |
| Blended price per 1M | $0.304 | $0.215 |
| Context window | 1M | 131K |
| Input / output per 1M | $0.188 / $0.652 | $0.12 / $0.5 |
| Max output | - | - |
| Input | text, vision | text |
| Released | 2025-04-05 | 2025-04-29 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | Llama 4 Maverick | Qwen3 30B-A3B |
|---|---|---|
| Coding | 2.2 Llama 4 Maverick | 3.0 Qwen3 30B-A3B |
| Agents | 9.9* Llama 4 Maverick | 11.1* Qwen3 30B-A3B |
| Reasoning | 19.4 Llama 4 Maverick | 19.8 Qwen3 30B-A3B |
| Knowledge & language | 10.5 Llama 4 Maverick | 9.9* Qwen3 30B-A3B |
| Human preference | 36.1 Llama 4 Maverick | 35.5 Qwen3 30B-A3B |
* Estimated: no published scores in that category. How the Respan Index works
Benchmarks both report
Llama 4 Maverick leads on 2 of the 3 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Llama 4 Maverick | Qwen3 30B-A3B |
|---|---|---|
| LMArena Elo | 1327 | 1327 |
| WeirdML (Håvard Tveit Ihle) | 24.5% | 29.8% |
| GPQA Diamond | 69.8% | 65.8% |
FAQ
Is Llama 4 Maverick better than Qwen3 30B-A3B?
Qwen3 30B-A3B scores 14.8 on the Respan Index (#151), ahead of Llama 4 Maverick at 14.3 (#152). Llama 4 Maverick leads on 2 of the 3 benchmarks both report.
Which is cheaper, Llama 4 Maverick or Qwen3 30B-A3B?
Llama 4 Maverick lists at $0.188 input and $0.652 output per 1M tokens; Qwen3 30B-A3B at $0.12 and $0.5. Qwen3 30B-A3B is 1.4x cheaper per token at list price.
Which has the larger context window?
Llama 4 Maverick accepts 1M tokens and Qwen3 30B-A3B 131K tokens.