Qwen3.8 Flash vs Gemini 3.8 Flash
Too close to call. On the work we tested, Qwen3.8 Flash and Gemini 3.8 Flash can't be told apart. Qwen3.8 Flash is the cheaper one, by 4.6×.
GrindBench-PubSec
CQwen3.8 FlashAlibaba (Qwen)49% 44–55%0.73¢ a task · 300 tasksCGemini 3.8 FlashGoogle52% 47–58%2.1¢ a task · 300 tasks
Too close to call here: their ranges overlap, so on another set of tasks the order could flip.
| Kind of work | Qwen3.8 Flash | Gemini 3.8 Flash |
|---|---|---|
| ATIP redaction | 2% | 18% |
| Procurement evaluation | 83% | 81% |
| Performance analysis | 53% | 33% |
| Bilingual equivalence | 53% | 97% |
GrindBench-Banking
CQwen3.8 FlashAlibaba (Qwen)52% 38–65%2.1¢ a task · 60 tasksCGemini 3.8 FlashGoogle48% 35–60%$0.11 a task · 60 tasks
Too close to call here: their ranges overlap, so on another set of tasks the order could flip.
| Kind of work | Qwen3.8 Flash | Gemini 3.8 Flash |
|---|---|---|
| Lending & mortgages | 50% | 50% |
| AML & KYC | 10% | 30% |
| Deposits & payments | 50% | 40% |
| Complaints & conduct | 100% | 60% |
| Credit risk | 60% | 60% |
| Capital & liquidity | 40% | 50% |
Each kind of work here rests on 10 tasks per model, so a single row can easily be 20 points off. A lead of 10 points or more is in bold; read it as a hint, not a result.