RA
Reka AI
Multimodal model · United States ·Mar 2025

Reka Flash 3

Multimodal Open (Restricted)
21B
Parameters

About Reka Flash 3

Reka Flash 3 is an AI model developed by Reka AI, in the multimodal category, released in 2025, made available as an open-weights model with 21B params.

On this page you'll find Reka Flash 3's full specifications. Review provider pricing and benchmark scores below, or compare Reka Flash 3 head-to-head with other multimodal models.

→ More models by Reka AI → All multimodal models → AI benchmarks leaderboard

Links

Announcement ↗ Hugging Face ↗

Run Reka Flash 3 locally — what to buy

Share this build ↗

Sized to what Reka Flash 3 actually needs (~8k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.

Good
Intel Arc A770 16GB
Consumer GPU · ~$329
~34.3 tok/s
Q4 · tight on context · up to 6k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →
Better
NVIDIA GeForce RTX 5070 Ti
Consumer GPU · ~$749
~66.3 tok/s
Q4 · tight on context · up to 6k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →
Best
NVIDIA GeForce RTX 5090
Consumer GPU · ~$1,999
~120 tok/s
Q4 · fits full context · up to 8k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →

Software support

·ExLlamaV2 ·KoboldCpp ·LM Studio ·MLX ·Ollama ·SGLang ·TensorRT-LLM ·llama.cpp ·vLLM ?KTransformers ?text-generation-webui

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Consumer GPU
13.7–120+ tok/s
22 options →
Apple
20.9–55.8 tok/s
17 options →
Datacenter GPU
56.6–120+ tok/s
13 options →

Community

Would you run Reka Flash 3 again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

More models like Reka Flash 3

AI
NeuroChess
· 72.3K
 RA
Reka Edge
Reka AI · 7B
 RA
Reka Core
Reka AI · 67B
 RA
Reka Flash
Reka AI · 21B
 Mistral AI
Mistral OCR 4
Mistral AI
 Tencent Hunyuan
Hy-Embodied-0.5-VLA
Tencent Hunyuan

Frequently asked questions

What is Reka Flash 3? ▶

Reka Flash 3 is an AI model developed by Reka AI, in the multimodal category, released in 2025. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created Reka Flash 3? ▶

Reka Flash 3 was developed by Reka AI and released in 2025.

Is Reka Flash 3 open source or proprietary? ▶

Reka Flash 3 ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.

How much does Reka Flash 3 cost? ▶

Pricing for Reka Flash 3 depends on the provider. See the providers table on this page for the latest API rates.

How does Reka Flash 3 perform on benchmarks? ▶

Benchmark scores for Reka Flash 3 are listed on this page as they are published. You can compare it head-to-head with other models on the GenAIList compare tool.

No videos yet. Know a good one? Add it below.

Know a great video about Reka Flash 3?