Mistral Saba vs Qwen3 VL 30B A3B Instruct
Side by side from live data: prices, limits, who serves each model and how reliably, and how they do on our chess arena.
In short
- Qwen3 VL 30B A3B Instruct is 1.1× cheaper at a typical mix of 3 input tokens per output token.
- Qwen3 VL 30B A3B Instruct accepts longer prompts: 262K vs 33K tokens.
- Qwen3 VL 30B A3B Instruct is the newer model (released 2025-10-06).
| Saba | Qwen3 VL 30B A3B Instruct | |
|---|---|---|
| Vendor | Mistral | Qwen |
| Released | 2025-02-17 | 2025-10-06 |
| Input / 1M | $0.20 | $0.15 |
| Output / 1M | $0.60 | $0.60 |
| Cached input / 1M | $0.02 | — |
| Blended / 1M (3:1) | $0.30 | $0.262 |
| Context | 33K | 262K |
| Max output | 26K | 16K |
| Accepts | text, file | text, image |
| Reasoning | No | No |
| Providers on OpenRouter | 3 | 3 |
| Cheapest provider, output | $0.60 | $0.60 |
| Best provider uptime, 24 h | 100% | 99.86% |
| Chess arena rating | not playing | not playing |
| Vendor API status | Operational | Operational |
Cost calculator
Saba
$21.00 per month$0.70 per day
Qwen3 VL 30B A3B Instruct
$18.00 per month$0.60 per day
Qwen3 VL 30B A3B Instruct saves $3.00 a month on this workload.
At list prices per 1M tokens; prompt caching and batch discounts are not included.
Compare Saba with
Compare Qwen3 VL 30B A3B Instruct with