We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.

New here? See how it works A 1-minute tour of the playground, models and your API key.

Mistral Large 4 vs Palmyra X5: API price, context & features · api14

Mistral: Mistral Large 4 vs Writer: Palmyra X5

A side-by-side comparison of Mistral: Mistral Large 4 ($2.45 / 1M tokens) and Writer: Palmyra X5 ($2.10 / 1M tokens) using current USD prices and capabilities from the api14 catalog. Both run through the same OpenAI-compatible API.

Specifications

	Mistral: Mistral Large 4	Writer: Palmyra X5
Provider	Mistral	Writer
Input	$2.45 / 1M tokens	$2.10 / 1M tokens
Output	$7.35 / 1M tokens	$21.00 / 1M tokens
Cache read	$0.25 / 1M tokens	-
Context window	1,048,576	1,040,000
Max output	262,144	8,192
Input	text, image	text
Output	text	text
Tool calling
Structured output
Reasoning control
Added	Oct 6, 2026	Jan 21, 2026

Key differences

Writer: Palmyra X5 is 14% cheaper on input tokens.
Mistral: Mistral Large 4 is 65% cheaper on output tokens.
Only Mistral: Mistral Large 4 supports tool calling.
Only Mistral: Mistral Large 4 supports structured JSON output.
Only Mistral: Mistral Large 4 supports reasoning effort control.
A request with 1,000 input and 500 output tokens costs $0.00613 with Mistral: Mistral Large 4 and $0.01 with Writer: Palmyra X5.

In depth

Mistral: Mistral Large 4 is made by Mistral and carries the model id mistralai/mistral-large-4-0 on api14. Writer: Palmyra X5 is made by Writer and carries the model id writer/palmyra-x5. Both are text category models, meaning they are built for language work rather than audio, video or embedding tasks, and both return text as their output. The practical difference starts with who stands behind them: Mistral is a model lab with a broad general purpose line, while Writer is a vendor whose models are aimed at enterprise writing and content workflows.

The two diverge most in what they accept and what you can control. Mistral Large 4 accepts both text and image input, so it can look at a screenshot, diagram or scanned page alongside your prompt, while Palmyra X5 accepts text only. Mistral Large 4 also exposes a much wider parameter surface: tools and tool_choice for function calling, structured_outputs and response_format for constrained JSON, reasoning and reasoning_effort plus include_reasoning for thinking style control, and sampling knobs such as frequency_penalty, presence_penalty, logprobs, top_logprobs and seed. Palmyra X5 supports a narrower set: max_tokens, stop, temperature, top_k and top_p. Their context windows are close enough that the difference is not meaningful in practice, roughly a million tokens each, so neither wins on how much you can feed in. On price, Palmyra X5 is the cheaper option for input, while Mistral Large 4 is the cheaper option for output.

For day to day selection, reach for Mistral Large 4 when your task needs anything beyond plain text in and text out. If you are building an agent that calls functions, if you need the response to conform to a JSON schema, if you want to tune reasoning effort, or if your input includes images, Palmyra X5 cannot do those things and Mistral Large 4 can. Reach for Palmyra X5 when the job is straightforward text generation or transformation, when your prompts are text only, and when input volume is the dominant cost in your workload, since it is the cheaper side on input. If your workload is output heavy, the cheaper output side is Mistral Large 4, which can offset its higher input rate.

Calling either model through api14 works the same way because the API is OpenAI compatible. You point your existing client at the api14 base URL, send a chat completions request with your messages, and change only the model field. Use mistralai/mistral-large-4-0 for Mistral Large 4 and writer/palmyra-x5 for Palmyra X5. Everything else in the request shape stays the same, so switching is a one line change and you can A/B the two on real traffic without rewriting your integration. Keep in mind that parameters are not universal: if you send tools, response_format or reasoning_effort to Palmyra X5, expect them to be unsupported, so gate those fields on the model id you are calling.

There are still cases where Palmyra X5 is the better pick even with the narrower feature set. If your prompts are pure text, you do not need tool calls or structured output, and input tokens dominate your bill, its cheaper input side makes it the more economical choice for high volume drafting, summarization or rewriting. Mistral Large 4 remains the safer default when you are unsure, because it covers the wider range of request types, but for a narrow text only pipeline Palmyra X5 is a reasonable and cheaper alternative.

Switch with one string

Both models use the same endpoint and request shape. Change the model value from "mistralai/mistral-large-4-0" to "writer/palmyra-x5" and prompts, streaming and tools keep working.

Mistral: Mistral Large 4 details & playground →Writer: Palmyra X5 details & playground →

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": "writer/palmyra-x5",
  "messages": [
    {
      "role": "user",
      "content": "Explain vector databases in two sentences."
    }
  ]
});
console.log(result.choices[0].message.content);

Mistral: Mistral Large 4 vs Writer: Palmyra X5

Specifications

Key differences

In depth

Switch with one string

More comparisons