Reka Flash 3
Summary
Reka Flash 3 is an open-weight language model from Rekaai, released on 10 Mar 2025. It accepts text and generates text, with a 66K-token context window and up to 59K output tokens per response. As of 19 Nov 2026, the lowest listed API price is $0.10 per million input tokens and $0.20 per million output tokens (OpenRouter), across 1 provider we track.
Efficient model for low-latency assistance, extraction, and routine automation
Last verified 19 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Rekaai |
|---|---|
| Release date | 10 Mar 2025 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | apache-2.0 [Hugging Face] |
| Parameters | 20.9B [Hugging Face] |
| Context window | 66K (65,536 tokens) [models.dev] |
| Max output | 59K tokens [models.dev] |
| Knowledge cutoff | 31 Jan 2025 [models.dev] |
| Input | Text [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | Yes [models.dev] |
| Tool calling | No [models.dev] |
| Structured output | Yes [models.dev] |
| Hugging Face | RekaAI/reka-flash-3 [Hugging Face] |
Reka Flash 3 API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| OpenRouter rekaai/reka-flash-3 | $0.10 | $0.20 | – | 66K | OpenRouter, 19 Nov 2026 |
Running Reka Flash 3 locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 13 GB | 15 GB | 18 GB |
| Q8_0 (8-bit) | 22 GB | 24 GB | 27 GB |
| FP16 / BF16 (16-bit) | 42 GB | 44 GB | 47 GB |
At 4-bit with an 8K conversation, Reka Flash 3 needs about 15 GB, so it fits on a graphics card with 16 GB or more, such as the NVIDIA GeForce RTX 4060 Ti 16 GB. Check your own hardware, or see Best GPU for local LLMs on our sister site AIPCs.
Frequently asked questions
What is the context window of Reka Flash 3?
Reka Flash 3 has a context window of 65,536 tokens (66K) and can generate up to 58,982 tokens in a single response, according to models.dev.
How much does the Reka Flash 3 API cost?
Reka Flash 3 is listed from $0.10 per million input tokens and $0.20 per million output tokens (OpenRouter). Prices last verified 19 Nov 2026.
Is Reka Flash 3 open source?
Reka Flash 3 is an open-weight model: its weights are published on Hugging Face as RekaAI/reka-flash-3 under the apache-2.0 license. Check the license terms before commercial use.
When was Reka Flash 3 released?
Rekaai released Reka Flash 3 on 10 Mar 2025, according to models.dev.
How much VRAM does Reka Flash 3 need?
About 15 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 24 GB at 8-bit. The 4-bit model fits on a graphics card with 16 GB or more, such as the NVIDIA GeForce RTX 4060 Ti 16 GB. These are estimates from the model's published parameter count and architecture.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| Gemma 3 27B IT | Google DeepMind | 12 Mar 2025 | 131K | $0.08 | $0.16 | Open |
| Mistral Small 3.1 24B | Mistral AI | 17 Mar 2025 | 128K | $0.10 | $0.30 | Open |
| Command R7B Arabic | Cohere | 27 Feb 2025 | 128K | $0.04 | $0.15 | Open |
| Gemma 3 12B IT | Google DeepMind | 12 Mar 2025 | 131K | $0.05 | $0.10 | Open |
| Llama Guard 4 12B | Meta | 29 Apr 2025 | 164K | $0.18 | $0.18 | Open |
| Phi-4-mini | Microsoft | 11 Dec 2024 | 128K | $0.07 | $0.30 | Open |
Change history
- Release date changed from 12 Mar 2025 to 10 Mar 2025