Model vs model

Gemini 4 Argon vs Mistral Small 4

Free to try, no account needed. You see each answer’s cost and speed.

At a glance

Longer bar means the bigger number.

Open or closed · can you download the model and run it yourself?

Gemini 4 ArgonClosed · only through Google’s API or a cloud that resells it
Mistral Small 4Open weights · download it, run it on your own hardware or any host

Blended price · per 1M tokens, lower is better

Gemini 4 Argon$4
Mistral Small 4$0.26

Intelligence Index · Artificial Analysis, higher is better

Gemini 4 Argon52.6
Mistral Small 411.3

Context window · tokens it can read at once

Gemini 4 Argon1.05M
Mistral Small 4262K

Where they sit among all 45 models

Gemini 4 Argon ranks 5th of the 41 models with an Intelligence Index, Mistral Small 4 38th. Up and to the left is smarter for less money. Hover or tap a dot for its numbers, then click it to compare that model with Gemini 4 Argon.

Show
Rank by
Intelligence: the Artificial Analysis Intelligence Index. Price: blended per 1M tokens, 3 parts input to 1 part output. Lines at $2 blended and the median score (38.9). The dashed line joins the best value models: nothing cheaper scores higher.

Which one to pick

Pick Gemini 4 Argon if

  • you want the higher overall score (52.6 vs 11.3)
  • you care about Terminal-Bench 4.0 (terminal tasks): 57.1 vs 0
  • you care about Humanity's Last Exam (hard reasoning): 57.1 vs 9.9
  • you need a longer context window (1.05M vs 262K tokens)

Pick Mistral Small 4 if

  • you want open weights you can download and run yourself
  • you want the lower bill ($0.26 vs $4 per 1M tokens blended)

See how they answer your prompt

Gemini 4 Argon’s “controversial” take on humans

We asked Gemini 4 Argon and 13 other models the same question three times each: “What is your most controversial take about humans? Give your own honest view, not a survey of opinions, and don't hedge.” Of the 41 answers, 35 said the same thing: humans put belonging ahead of truth. Gemini 4 Argon gave it all three times. Mistral Small 4 wasn’t in that test. Read the full story.

Gemini 4 Argon

Its first take, as a cartoon it described and Nano Banana 2.1 drew.
  1. Answer 1 · the common take“Humans did not evolve to find truth; they evolved to win arguments and secure status within their tribes.”
  2. Answer 2 · the common take“Humans are far less rational than they believe, using their advanced intellect primarily to invent complex justifications for decisions already made by primitive instincts.”
  3. Answer 3 · the common take“Human consciousness is largely a PR department designed to invent noble justifications for base evolutionary impulses, rather than the actual decision-maker it claims to be.”

Answers exactly as the models wrote them. “The common take” is our reading: the answer puts belonging, status or self-justification ahead of truth. How we ran it

Ask both the same question

How smart is each one?

Of the 12 scores both models have, Gemini 4 Argon is ahead on every one.

Every intelligence score we have

Out of 100, higher is better. Gemini 4 Argon Mistral Small 4

Overall

  • Intelligence Index

By area

  • Coding

    Gemini 4 Argon: not scored yet

  • Agentic work

    Gemini 4 Argon: not scored yet

By field of work

  • Finance & accounting

  • Legal

  • Healthcare & medicine

  • Economics

  • Engineering

  • Strategy & operations

Benchmarks

  • Factual knowledge · SimpleQA

    Mistral Small 4: not scored yet

  • Hard reasoning · Humanity's Last Exam

  • Science · GPQA Diamond

    Gemini 4 Argon: not scored yet

  • Competition math · OTIS Mock AIME

    Mistral Small 4: not scored yet

  • Research math · FrontierMath

    Mistral Small 4: not scored yet

  • Scientific coding · SciCode

  • Coding in Cursor · CursorBench 4.0

    Mistral Small 4: not scored yet

  • Mergeable code · FrontierCode 2.0

    Mistral Small 4: not scored yet

  • Terminal tasks · Terminal-Bench 4.0

  • Tool use · τ-Bench Banking

    Gemini 4 Argon: not scored yet

  • Instruction following · IFBench

    Gemini 4 Argon: not scored yet

  • Long documents · AA-LCR

  • Uncensored · UGI

Our tests

  • Next.js coding tasks · Vercel's Next.js evals, % of tasks passed

    Mistral Small 4: not scored yet

Test them on your own question

The specs

Spec Gemini 4 Argon Mistral Small 4
Input price$2 per 1M$0.15 per 1M
Output price$10 per 1M$0.60 per 1M
Cost per test task$1.99$0.0150
Context window1.05M262K
Max output33K256K
AcceptsText, Image, Video, Audio, PDFText, Image, PDF
Open weightsNoYes
LabGoogleMistral AI
ReleasedOct 2026Mar 2026
Knowledge cutoffMar 2026Jun 2025

Blended price weighs 3 parts input to 1 part output. List prices from models.dev. Cost per test task from Artificial Analysis.

Questions

Is Gemini 4 Argon or Mistral Small 4 cheaper?

Mistral Small 4 is cheaper. It charges $0.15 per 1M input tokens and $0.60 per 1M output tokens, against $2 and $10 for Gemini 4 Argon. Blended 3 parts input to 1 part output, that is $0.26 against $4 per 1M tokens.

Are Gemini 4 Argon and Mistral Small 4 open source?

Only Mistral Small 4. It is an open-weights model: Mistral AI publishes the weights, so you can download it, run it on your own hardware or pick any host. Gemini 4 Argon is closed: you can only use it through an API.

Which is smarter, Gemini 4 Argon or Mistral Small 4?

Gemini 4 Argon scores higher on the Artificial Analysis Intelligence Index, 52.6 against 11.3 for Mistral Small 4. The biggest differences are on Terminal-Bench 4.0 (57.1 vs 0) and Humanity's Last Exam (57.1 vs 9.9).

Which has the longer context window?

Gemini 4 Argon accepts 1.05M tokens, against 262K for Mistral Small 4.

Can I compare Gemini 4 Argon and Mistral Small 4 on my own prompt?

Yes. ModelCompare sends one prompt to both models at once and shows each answer side by side with what it cost. You can try it free without an account.

Still deciding? Send one prompt to both and read the answers side by side, with what each one cost. Compare Gemini 4 Argon and Mistral Small 4 →

Where the numbers come from

Intelligence Index and benchmark scores are from Artificial Analysis, at each model’s highest effort, except the Epoch Capabilities Index, SimpleQA, OTIS Mock AIME and FrontierMath, which are from Epoch AI, and the Uncensored score, which is from the UGI Leaderboard, and CursorBench, which is from Cursor, and FrontierCode, which is from Cognition at each model’s best-scoring effort. The hallucination test is ours: questions with one exact answer, fake evidence pushed against right answers and requests that tempt a model to name packages that don’t exist. Prices are list prices from models.dev. Example answers come from our own test runs. Updated November 13, 2026.