Dots Studio: Dots3-Note Preview (free) vs Recursive: Flint (free)
A side-by-side comparison of Dots Studio: Dots3-Note Preview (free) (Free) and Recursive: Flint (free) (Free) using current USD prices and capabilities from the api14 catalog. Both run through the same OpenAI-compatible API.
Specifications
| Dots Studio: Dots3-Note Preview (free) | Recursive: Flint (free) | |
|---|---|---|
| Provider | Dots Studio | Recursive |
| Context window | 512,000 | 262,144 |
| Max output | 460,800 | 235,929 |
| Input | text, image | text |
| Output | text | text |
| Tool calling | ||
| Structured output | ||
| Reasoning control | ||
| Added | Aug 14, 2026 | Oct 21, 2026 |
Key differences
- Dots Studio: Dots3-Note Preview (free) reads 2.0× more context (512,000 tokens).
- Only Dots Studio: Dots3-Note Preview (free) supports structured JSON output.
In depth
Dots Studio: Dots3-Note Preview (free) is a text model from Dots Studio, listed under the text category and released by the provider on 2026-08-14. Recursive: Flint (free) is a text model from Recursive, also in the text category, with a provider release date of 2026-10-21. Both are offered as free variants on api14, so neither carries a price advantage over the other based on the information available. The two differ mainly in how much context they accept, what input they can read, and which request parameters they honour.
Dots3-Note Preview has the larger context window of the two, at 512000 tokens against 262144 for Flint. It also accepts image input alongside text, while Flint accepts text only; both produce text output. On structured output, Dots3-Note Preview is the one that supports it, so if you need the model to return data conforming to a schema, that is the side to use. Both models expose reasoning controls (include_reasoning and reasoning) and both support tools, so function calling is available on either. Flint adds a few sampling and inspection parameters that Dots3-Note Preview does not list, namely logprobs, top_logprobs, seed and stop. The qualitative comparison marks the larger context as belonging to Dots3-Note Preview and structured output as unique to it, with no meaningful difference recorded for tool support or reasoning.
For practical selection, reach for Dots3-Note Preview when the job involves long inputs, such as large documents, extended conversation history or big code files, since its context window is roughly double. Pick it as well when you need to send images together with text, or when you want the response constrained to a JSON schema through structured outputs. Flint is the better fit when you want deterministic sampling via seed, or when you need token level detail through logprobs and top_logprobs, which are useful for classification confidence, evaluation harnesses and debugging odd generations. If your task is plain text in and text out, fits comfortably inside 262144 tokens, and does not need schema enforcement, either model will do the job and the choice comes down to which extra parameters matter to you.
Calling either model through api14 works the same way, because the gateway exposes an OpenAI-compatible API. You point your existing OpenAI client at the api14 base URL, supply your api14 key, and set the model field to the id you want. To use Dots3-Note Preview, set the model to "dots-studio/dots-3-note-preview:free". To use Flint, set it to "recursive/flint-v1.0:free". Everything else in the request body, messages, temperature, max_tokens, tools and the rest, keeps the same shape, so switching between the two is a one line change to the model id rather than a rewrite of your integration. Parameters that a given model does not list, such as seed on Dots3-Note Preview or structured_outputs on Flint, should simply be omitted for that model.
There are cases where the smaller model is still the right call. If you depend on seed for reproducible outputs, or on logprobs and top_logprobs to inspect how the model distributed probability across tokens, Flint is the only one of the two that lists those parameters, so it remains the better pick regardless of context size. The same applies if you are already running Flint in production and your inputs fit within its context window, since there is no capability in Dots3-Note Preview that you would strictly need for text only work without images or schema constraints.
Switch with one string
Both models use the same endpoint and request shape. Change the model value from "dots-studio/dots-3-note-preview:free" to "recursive/flint-v1.0:free" and prompts, streaming and tools keep working.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.chat.completions.create({
"model": "recursive/flint-v1.0:free",
"messages": [
{
"role": "user",
"content": "Explain vector databases in two sentences."
}
]
});
console.log(result.choices[0].message.content);
More comparisons
- LiquidAI: LFM3-1.7B vs Ricursive: Ripple 65B Preview
- LiquidAI: LFM3-1.7B vs Crouton Alpha
- LiquidAI: LFM3-1.7B vs Tencent: Hy4
- LiquidAI: LFM3-1.7B vs Z.ai: GLM 5.3 FlashX Legal
- Ricursive: Ripple 65B Preview vs Crouton Alpha
- Ricursive: Ripple 65B Preview vs Tencent: Hy4
- Ricursive: Ripple 65B Preview vs Z.ai: GLM 5.3 FlashX Legal
- Crouton Alpha vs Tencent: Hy4