Model Terminal
Reka Flash 3
Reka Flash 3 is a 21B-parameter open-weight reasoning LLM from Reka AI, designed for chat, coding, instruction following, and function calling with explicit reasoning tags and a focus on efficiency and low-latency deployment. Source: supabase.
Value score
Not scored
Context
—
tokens
Max output
—
tokens
Price
—
Capability radar
Peer value bars
Identity
- Developer
- Reka AI
- Openness
- —
- Modalities
- —
- Release
- —
- Knowledge cutoff
- —
- Deprecation
- —
- API docs
- —
Benchmarks
No benchmark scores yet.
Compare nearby
Pricing
- Input / 1M
- —
- Output / 1M
- —
- Speed
- —
