Flint (free)
Flint (free) is a large language model made by Recursive, added to the catalog on 21 Oct 2026. It has a context window of 262,144 tokens and has no per-token charge. It accepts text and produces text. Typical uses: coding-agent workflows, working with long code or text contexts and tool-assisted coding tasks.
Price checked ยท source: OpenRouter
What Flint (free) is
Recursive Flint is an open-weight mixture-of-experts model from Recursive, described as built for coding agents. It accepts text and produces text. Its supported capabilities include reasoning, tool use, and response-format controls. These features align with its stated focus on coding-agent work, while its listed input and output modalities are text. The catalog does not specify programming languages, particular tools, or named output formats, so its scope is described through these general capabilities rather than a specific software stack.
Flint has a context window of 262,144 tokens and a maximum output of 235,929 tokens. This gives it room to handle substantial text as input and return extended text, subject to the listed limits. Its context size may suit tasks that involve keeping lengthy prompts, code, or related text in context; the catalog does not identify specific repository features or promise outcomes. Tool support is relevant to agent workflows that use tools, while response-format support can be useful when a task calls for a specified output structure. Reasoning is also listed as a supported capability. The provided details do not name supported formats or tool integrations. Flint is a text-in, text-out option for coding agents and other text tasks that benefit from a large context and its supported reasoning, tools, and response-format features.
Key features
- Accepts text and produces text
- Has a 262,144-token context window
- Supports output of up to 235,929 tokens
- Supports tools, reasoning, and response formats
- Described as an open-weight mixture-of-experts model
Typical uses
- Coding-agent workflows
- Working with long code or text contexts
- Tool-assisted coding tasks
- Generating text in specified response formats
Flint (free) pricing
US dollars, list price through OpenRouter.
| Item | Free |
|---|---|
| Input, per 1M tokens | $0 |
| Output, per 1M tokens | $0 |
No price change since it was added on 21 Oct 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0 |
| Summarise a 50-page PDF25,000 in / 800 out | $0 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0 |
| Read a 300-page contract set150,000 in / 1,500 out | $0 |
Flint (free) specifications
| Model id | recursive/flint-v1.0:free |
|---|---|
| Version | flint-v1.0-20261021 |
| Context window | 262,144 tokensAbout 196,608 English words |
| Max output | 235,929 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 21 Oct 2026 |
| Input types | text |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | No |
| 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 Flint (free) yet.