Nemotron 3 Ultra vs Qwen3.8 Flash: cost per task
Qwen3.8 Flash is cheaper on 13 of 14 tasks and about 5.5x cheaper on average (geometric mean).
| Task | Nemotron 3 Ultra | Qwen3.8 Flash | Difference |
|---|---|---|---|
| Answer a chat question | 0.33¢ | 0.08¢ | 4.3x |
| Reply to a customer email | 0.86¢ | 0.04¢ | 23.3x |
| Write a 1,500-word blog post | 1.9¢ | 0.75¢ | 2.5x |
| Write 10 product descriptions | 0.75¢ | 0.76¢ | 1.0x |
| Translate 1,000 words | 3.0¢ | 0.08¢ | 36.4x |
| Proofread one page (500 words) | 1.0¢ | 0.10¢ | 10.6x |
| Summarize a 20-page PDF | 1.3¢ | 0.32¢ | 4.1x |
| Extract data from an invoice | 0.66¢ | 0.11¢ | 6.0x |
| Analyze a 200-row spreadsheet | 2.8¢ | 0.85¢ | 3.3x |
| Answer from 5 retrieved documents (RAG) | 0.99¢ | 0.19¢ | 5.2x |
| Classify 100 support tickets | 1.8¢ | 0.42¢ | 4.4x |
| Review 500 lines of code | 2.5¢ | 0.83¢ | 3.0x |
| Write a SQL query | 0.76¢ | 0.13¢ | 6.0x |
| Run a 50-step coding agent | 67¢ | 9.2¢ | 7.2x |
Specs and list prices
| Nemotron 3 Ultra | Qwen3.8 Flash | |
|---|---|---|
| Input per 1M | $0.60 | $0.15 |
| Output per 1M | $2.40 | $0.47 |
| Cached input per 1M | $0.12 | $0.016 |
| Context | 262K | 1M |
| Providers | 5 | 2 |
| Reasoning | default | none |
| LMArena Text | 1426 | - |
| Released | Jun 4, 2026 | Aug 26, 2026 |
FAQ
Is Nemotron 3 Ultra or Qwen3.8 Flash cheaper?
Qwen3.8 Flash is cheaper on 13 of 14 tasks and about 5.5x cheaper on average (geometric mean).
How much does Nemotron 3 Ultra cost compared to Qwen3.8 Flash per token?
Nemotron 3 Ultra: $0.60 input and $2.40 output per 1M tokens. Qwen3.8 Flash: $0.15 input and $0.47 output. Token prices alone mislead - models use different numbers of tokens for the same task, which is what the task table shows.