Reka Edge
VerifiedReka Edge processes images, text, and video in a single multimodal workflow.
About Reka Edge
Reka Edge combines vision and language capabilities into one system. Its architecture handles simultaneous inputs from images, video frames, and text prompts. The 16384-token context allows processing of longer multimodal sequences without truncation.
As a closed-weight model, Reka Edge is accessed through hosted APIs rather than local deployment. It is designed for scenarios requiring unified understanding of visual and textual data streams. Typical applications include video summarization, image-grounded question answering, and cross-modal content generation.
Capabilities
How Reka Edge compares
Reka Edge (striped bar) vs other multimodal on intelligence, speed and price.
Price
USD per 1M output tokens · Lower is better · Reka Edge ranks #1 of 67
Sources: Artificial Analysis (intelligence, speed) · OpenRouter (price).
Best for
Video Understanding Tasks
Reka Edge excels at multimodal analysis of video content combined with text, enabling detailed visual interpretation and conditioned text generation from dynamic inputs.
Image-Text Cross-Modal Projects
The model performs strong cross-modal reasoning between images and text, supporting tasks that require integrated visual analysis and contextual text responses.
Extended Multimodal Contexts
With context handling up to 16384 tokens, it suits workflows involving lengthy documents, image sequences, or video transcripts for sustained multimodal interactions.
Strengths & limitations
Strengths
- +Native support for video alongside image and text
- +Unified multimodal architecture
- +Practical context length for many tasks
Limitations
- –Context window smaller than many frontier models
- –Edge-oriented design may limit peak capability
- –Performance details not widely benchmarked publicly
Pricing by provider
Live per-provider pricing & uptime, routed via OpenRouter. Prices are USD per 1M tokens.
| Provider | Input /1M | Output /1M | Context | Uptime |
|---|---|---|---|---|
| Reka(bf16) | $0.10 | $0.10 | 16K | — |
Cost calculator
Estimate what Reka Edge would cost for your usage.
Based on Reka Edge's $0.10/1M input · $0.10/1M output. Estimate only — actual cost varies by provider and caching.
Quick start
OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. Only the model slug changes between models.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://openrouter.ai/api/v1",
apiKey: process.env.OPENROUTER_API_KEY,
});
const completion = await client.chat.completions.create({
model: "rekaai/reka-edge",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(completion.choices[0].message.content);Model slug: rekaai/reka-edge
Editor's verdict
Reka Edge is Rekaai's proprietary multimodal with a 16K-token context window.
At $0.10 per 1M output tokens, it is very cost-efficient for its class, served by 1 provider.
It is available through Rekaai's API and aggregators like OpenRouter.
Best suited to native support for video alongside image and text and unified multimodal architecture.
Frequently asked questions
Reka Edge handles context up to 16384 tokens for processing extended multimodal inputs.
User reviews
Real, verified reviews from the community shape this model's rating.
Loading reviews…
Similar models
Other multimodal worth comparing.