Reka: Flash 3
REKAAI Developer Architecture Profile
Intelligence (ELO)1432Chatbot Arena Verified
Max Context65,536Tokens
API Cost / 1M$0.30Blended Prompt + Completion
Model Capabilities
- Drafting
- Classification
- Coding & Logic
- Fictional
Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka. It excels at general chat, coding tasks, instruction-following, and function calling. Featuring a 32K context length and optimized through reinforcement learning (RLOO), it provides competitive performance comparable to proprietary models within a smaller parameter footprint. Ideal for low-latency, local, or on-device deployments, Reka Flash 3 is compact, supports efficient quantization (down to 11GB at 4-bit precision), and employs explicit reasoning tags ("<reasoning>") to indicate its internal thought process.
Reka Flash 3 is primarily an English model with limited multilingual understanding capabilities. The model weights are released under the Apache 2.0 license.
Granular Pricing Matrix
Input Tokens (Prompt)$0.10 / 1M
Output Tokens (Completion)$0.20 / 1M
Pricing data via OpenRouter. Sync: 3/31/2026
Evaluate Competitors
VS Engine MatchupReka: Flash 3 vs NVIDIA: Nemotron 3 SuperVS Engine MatchupReka: Flash 3 vs NVIDIA: Nemotron Nano 12B 2 VLVS Engine MatchupReka: Flash 3 vs TheDrummer: Cydonia 24B V4.1VS Engine MatchupReka: Flash 3 vs Tencent: Hunyuan A13B InstructVS Engine MatchupReka: Flash 3 vs TNG: DeepSeek R1T2 ChimeraVS Engine MatchupReka: Flash 3 vs xAI: Grok 3 Mini