Qwen3.7 Flash
by Alibaba (Qwen) · qwen/qwen3.7-flash · #14 cheapest of 422 paid models
Prices updated Sep 19, 2026, 6:22 PM UTC · refreshed hourly
Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer interaction, with strengths in object recognition, spatial understanding, and real-world...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.03
per 1M tokens
Output
$0.13
per 1M tokens
Blended (3:1)
$0.055
3 input : 1 output
Cache read
$0.0060
per 1M cached tokens
Cache write
$0.038
per 1M tokens
Cache & batch economics
Effective prices for Qwen3.7 Flash with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.055 | $0.00319 | — |
| 50% | $0.046 | $0.00199 | −38% |
| 90% | $0.039 | $0.00103 | −68% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $5.40e-5 | $0.054 |
| Document summary | 8,000 | 1,000 | $0.00037 | $0.37 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.00116 | $1.16 |
| Long-context analysis | 150,000 | 5,000 | $0.00515 | $5.15 |
Specs
Context window
1M
tokens
Max output
66K
tokens
Input modalities
text, image, video
Tokenizer
Qwen