Reka Flash 3

reasoningapache-2.0

Reka Flash 3 is a 21B general-purpose reasoning LLM which excels at chat, coding, and instruction-following while maintaining a level of performance competitive with significantly larger models. Note: Reka Flash 3 is primarily an English model with limited multilingual understanding capabilities.

Context
66K
Max output
66K
Input price
$0.1/1M tokens
Output price
$0.2/1M tokens

Capabilities

Reasoning & Interaction
Reasoning
Output Control
Structured Output
JSON Mode
Infrastructure
Streaming
Fine-tuning

Details

Model IDreka-flash-3
Provider Reka AI
Creator reka
Familyreka-flash
Licenseapache-2.0
Statusactive
Typereasoning
Input modalitiestext
Output modalitiestext
Open weightYes
Parameters—
Architecture—
Context window66K tokens
Max input—
Max output66K tokens
Reasoning tokensYes
Knowledge cutoff—
Release date
Sourceofficial

Tools

Function Callingfunction_calling
Call external functions and APIs

Endpoints

Chat CompletionsPOST
Generate chat responses with messageshttps://api.reka.ai/v1/chat/completions

Pricing

Input
$0.1
Output
$0.2
Cache write
—
Cache read
—
Batch in
—
Batch out
—

Family Comparison: reka-flash

ModelContextMax outPricing
Reka Flash 366K66K$0.1$0.2
Reka Flash128K—$0.8$2

API

GET/v1/models/reka/reka-flash-3