Qwen3.6 35B A3B vs Nemotron 3.5 Lightning
Too close to call. On the work we tested, Qwen3.6 35B A3B and Nemotron 3.5 Lightning can't be told apart. Nemotron 3.5 Lightning is the cheaper one, by 2.2×.
GrindBench-PubSec
DQwen3.6 35B A3BAlibaba (Qwen)33% 27–38%0.53¢ a task · 300 tasksDNemotron 3.5 LightningNVIDIA23% 19–28%0.19¢ 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 | Qwen3.6 35B A3B | Nemotron 3.5 Lightning |
|---|---|---|
| ATIP redaction | 0% | 0% |
| Procurement evaluation | 58% | 42% |
| Performance analysis | 11% | 16% |
| Bilingual equivalence | 85% | 42% |
GrindBench-Banking
FQwen3.6 35B A3BAlibaba (Qwen)3% 0–8%1.8¢ a task · 60 tasksFNemotron 3.5 LightningNVIDIA2% 0–5%0.87¢ 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 | Qwen3.6 35B A3B | Nemotron 3.5 Lightning |
|---|---|---|
| Lending & mortgages | 20% | 10% |
| AML & KYC | 0% | 0% |
| Deposits & payments | 0% | 0% |
| Complaints & conduct | 0% | 0% |
| Credit risk | 0% | 0% |
| Capital & liquidity | 0% | 0% |
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.