vakati-tech.ai

// head_to_head

Mistral Large 4 vs Pareto 26.10 Preview

Published LiveBench scores across all seven categories, live list pricing, context windows, and the measured cost of a point of capability — for both models, side by side.

OpenRouter + LiveBenchAll comparisonsFull leaderboard

Mistral Large 4 is the cheaper of the two; neither can be ranked on quality here.

Pareto 26.10 Preview does not have a published LiveBench run, so this comparison covers price, context and declared capabilities only. A missing score means "not evaluated", not "worse" — the right way to separate these two is an eval on your own workload.

mistralai

Mistral Large 4

Blended / 1M
$1.05
Context
1.0M
Released
Oct 6, 2026
Overall score
40.9
reasoningtool callingimage inputprompt caching

unbiased

Pareto 26.10 Preview

Blended / 1M
$1.40
Context
1.0M
Released
Oct 1, 2026
Overall score
Not evaluated
tool callingimage inputprompt caching

Specs and pricing

MetricMistral Large 4Pareto 26.10 Preview
LiveBench overall

Weighted mean of the seven LiveBench category scores, 0–100. Higher is better.

40.9—
Cost per point

Measured benchmark spend divided by overall score — dollars per point of capability.

$0.0104—
Blended price / 1M

3:1 input:output mix, the usual shape of production traffic.

$1.05win$1.40
Input price / 1M$0.700win$0.800
Output price / 1M$2.10win$3.20
Cached input / 1M

Price of an input token served from the prompt cache, where the provider publishes one.

$0.070$0.030win
Context window1.0M1.0M
Max output tokens262Kwin131K

Benchmarks by category

An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Mistral Large 4 on top, Pareto 26.10 Preview below, both out of 100.

Long-horizon agentic
9.2
—
Coding
44.4
—
Reasoning
83.9
—
Browser
0.6
—
Data analysis
76.5
—
Research
30.8
—
Documents & design
60.7
—

What each one costs to run

Per-token prices are hard to feel. These are monthly list costs for both models across five workload shapes, using each provider's published cached-input rate where there is one.

WorkloadMistral Large 4Pareto 26.10 Preview
Support chatbot

1.2K in / 400 out × 200K requests

$290.64/mo$392.56/mo
RAG assistant

8K in / 600 out × 100K requests

$434.00/mo$524.00/mo
Coding agent

40K in / 4K out × 20K requests

$375.20/mo$464.80/mo
Document extraction

20K in / 1.5K out × 50K requests

$826.00/mo$1,002/mo
Bulk classification

500 in / 20 out × 5M requests

$1,645/mo$1,935/mo
Run these two through the cost calculator

Which should you pick?

You are cost-constrained

Mistral Large 4

Cheaper on blended list price at $1.05 per million tokens.

Mistral Large 4 vs Pareto 26.10 Preview FAQ

Which is better, Mistral Large 4 or Pareto 26.10 Preview?

Mistral Large 4 is the cheaper of the two; neither can be ranked on quality here. Pareto 26.10 Preview does not have a published LiveBench run, so this comparison covers price, context and declared capabilities only. A missing score means "not evaluated", not "worse" — the right way to separate these two is an eval on your own workload.

Is Mistral Large 4 cheaper than Pareto 26.10 Preview?

Mistral Large 4 is cheaper. On a 3:1 input:output blend, Mistral Large 4 lists at $1.05 per million tokens and Pareto 26.10 Preview at $1.40 — Mistral Large 4 is 33% cheaper. Input and output are priced separately — Mistral Large 4 charges $0.700 in and $2.10 out, Pareto 26.10 Preview charges $0.800 and $3.20 — so the model that looks cheaper flips depending on how output-heavy your workload is.

Does Mistral Large 4 or Pareto 26.10 Preview have a bigger context window?

They are effectively the same — 1.0M for Mistral Large 4 and 1.0M for Pareto 26.10 Preview.

Do Mistral Large 4 and Pareto 26.10 Preview support prompt caching?

Both publish a cached-input rate: $0.070 per million for Mistral Large 4 and $0.030 for Pareto 26.10 Preview, against full input rates of $0.700 and $0.800. On a workload with a long stable prefix — a system prompt, a tool schema, a retrieved corpus — that changes the economics more than the headline price does.

Related comparisons

How these numbers are produced

  • Price — provider list price from OpenRouter, refreshed every 15 minutes. “Blended” is a 3:1 input:output mix.
  • Scores — LiveBench release 2026-11-06, using their own category map. Each model shows its strongest published run. A blank means “not evaluated”, never “bad”.
  • Cost per point — the measured dollars LiveBench spent on the run, divided by the score it earned.
  • “Win” — awarded only past a threshold: one full point on a benchmark score, 10% on a price, 25% on a context window. Anything tighter reports as a tie, because effort settings alone move a LiveBench score by more than that.

Published benchmarks rank models on someone else's tasks. Before committing, see LLM & agent evaluation for building an eval on your own.