model · class value · updated 2026-07-13
What it's for
Best for: Local visual perception, image/question tasks, and OCR-style workflows.
Not ideal for: Frontier multimodal reasoning.
Role: Small local VLM
Specifications
| Provider | Alibaba |
| API format | openai |
| Context window | 256000 |
| Restricted access | No |
| Capabilities | Vision, Tools |
| Execution Paths | api, self_host |
Price components
| Type | USD | Basis | Source | Verified | Freshness |
|---|---|---|---|---|---|
| per_million_in | $0.12 | required · verified | OpenRouter + src_model_price_api | 2026-07-13 | vendor · 1d |
| per_million_out | $0.46 | required · verified | OpenRouter + src_model_price_api | 2026-07-13 | vendor · 1d |
Cost at three usage tiers
| A few questions and small tasks each day. | $1.11/mo |
| Daily use, plus some scheduled tasks. | $3.71/mo |
| Working for you most of the day, including web tasks. | $11.12/mo |
Token cost only; add a host for the full-stack estimate.
Compare