Reka Edge
Reka Edge
Seven billion parameters that see, at a price that rounds to nothing.
Updated
The read
Reka Edge is a small vision-language model priced flat and very low, which makes it the practical choice for image-heavy pipelines: captioning, moderation, tagging, extracting text from a photograph. It is efficient enough to run at volume without a budget conversation. The context window is small, the reasoning is shallow, and it is a specialist rather than an assistant.
Where it shines
- Among the cheapest multimodal models available
- Efficient enough to run across a whole media library
- Flat input and output pricing, so cost is easy to predict
Worth knowing first
- Very small context window
- Not an assistant: general conversation is beyond it
Index score
68.7
Provisional. This score is read from the product's public capability, not from the one-prompt rebuild the fully reviewed entries went through, so treat it as a placing rather than a verdict.
- Rank
- 72 of 73
- Builds
- LLMs
- Category
- Small and fast
- Output
- Images and text in, text out
- Pricing
- Paid only
- Best for
- High-volume image captioning, tagging and moderation
Scores are our own editorial judgement, weighted craft 55, speed 10, control 15 and value 20.
Also worth a look
Full index- 11Claude Haiku 5.5Claude Haiku 5.5ProvisionalA small model priced like a rounding error that scores where mid-tier models did last quarter.Small and fastPaid only87.9
- 22GPT-6 LunaGPT-6 LunaProvisionalThe cheap end of GPT-6, and the model free ChatGPT users now talk to.Small and fastPaid only84.9
- 30GPT-5.6 LunaGPT-5.6 LunaProvisionalStill cheap and still capable, but GPT-6 Luna costs half as much.Small and fastPaid only83.0
