Reka Flash 3
Reka Flash 3 is a large language model made by Rekaai, added to the catalog on 12 Mar 2025. It has a context window of 65,536 tokens and costs $0.10 per 1M input tokens and $0.20 per 1M output tokens. It accepts text and produces text. Typical uses: general chat, coding tasks and following written instructions.
Price checked ยท source: OpenRouter
What Reka Flash 3 is
Reka Flash 3 is a general-purpose, instruction-tuned large language model developed by Reka. It has 21 billion parameters and accepts text input to produce text output. The catalog describes it for general chat, coding tasks, instruction following, and function calling. It also supports reasoning and structured outputs. These capabilities make it relevant to conversational exchanges, code-related requests, and tasks where responses need to follow explicit directions or a defined format. It is described as general-purpose rather than tied to a particular subject area.
The model has a context length of 65,536 tokens and a maximum output of 58,982 tokens. The context limit indicates how much text it can take into account in a request, while the output limit sets the ceiling for a generated response. This context capacity suits work involving substantial supplied passages, such as asking questions about documents, summarising material, or carrying instructions across a longer prompt. The output allowance can accommodate extended explanations, code drafts, or structured text when a task calls for them. Its text-only input and output make it suitable for workflows built around written prompts and written results, rather than image or other media input. The listed support for reasoning, structured outputs, and function calling may suit requests that combine multi-step text work with formatted results or a function-call workflow. The catalog does not specify particular tools or output schemas.
Key features
- Accepts text and produces text.
- Has a 65,536-token context window.
- Supports a maximum output of 58,982 tokens.
- Supports reasoning and structured outputs.
- Described for chat, coding, instruction following, and function calling.
Typical uses
- General chat
- Coding tasks
- Following written instructions
- Function-calling workflows
- Producing structured text
Reka Flash 3 pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.10 |
| Output, per 1M tokens | $0.20 |
No price change since it was added on 12 Mar 2025.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0001 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0027 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0034 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.015 |
Reka Flash 3 specifications
| Model id | rekaai/reka-flash-3 |
|---|---|
| Version | reka-flash-3 |
| Context window | 65,536 tokensAbout 49,152 English words |
| Max output | 58,982 tokens |
| Knowledge cutoff | 2025-01-31 |
| Added to catalog | 12 Mar 2025 |
| Input types | text |
|---|---|
| Output types | text |
| Tool calling | No |
| Structured outputs | Yes |
| Reasoning effort control | No |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Reka Flash 3 yet.