Comparison · Mistral

Mistral Large 4 vs Mistral Small 4

Listed API prices of Mistral Large 4 and Mistral Small 4 turned into the cost of the same three standard tasks. Output quality, speed and answer length are not compared.

Calculated impact cost per task and per month

For an L2 task (deep analysis over a 50,000-token document, 3 steps), Mistral Large 4 costs $0.118 and Mistral Small 4 $0.026 at the prices listed in the United States on , so Mistral Large 4 is 358.5 % more than Mistral Small 4, counting visible output only, without cache discounts or hidden reasoning.

Head to head: Mistral Large 4: $0.118; Mistral Small 4: $0.026 Calculated from published rates Input and visible output are priced separately for the same task. 159,000 input + 3,000 visible output tokens per task. No cache discount; hidden reasoning excluded.
Head to head.Calculated from published rates Input and visible output are priced separately for the same task. 159,000 input + 3,000 visible output tokens per task. No cache discount; hidden reasoning excluded. Open image · Explore the chart · Measured Lab · CC BY 4.0
TaskMistral Large 4, per taskMistral Small 4, per taskMistral Large 4, per monthMistral Small 4, per monthMistral Large 4 vs Mistral Small 4, per task
L1Micro-task (classification or strict JSON extraction)$0.0015per L1 task$0.0003per L1 task$21.46per month$5.04per month+326.1 %
L2Deep analysis over a 50,000-token document$0.118per L2 task$0.026per L2 task$1,716.96per month$374.49per month+358.5 %
L3Agentic coding loop, 15 steps$0.940per L3 task$0.203per L3 task$13,720.35per month$2,959.78per month+363.6 %

Calculated Per month: 20 tasks an hour around the clock, 730 hours.

Prices USD per 1M tokens

PriceMistral Large 4Mistral Small 4
Input, per 1M tokens$0.700$0.150
Output, per 1M tokens$2.10$0.600
Price fromProvider's price list · source page PublishedOpenRouter · source page (data) Published
Read · the United States · the United States
Context window1,048,576 tokens262,144 tokens

What can I use it for? six common tasks, our data checks

Quality on your task: not measured by Measured Lab yet.

TaskMistral Large 4Mistral Small 4
Answer customer messagesFits on price and contextFits on price and context
Pull data from invoices and documents with strict JSON and page imagesFits on price and contextFits on price and context
Summarize documents and meetingsFits on price and contextFits on price and context
Translate contentFits on price and contextFits on price and context
Chat assistant on your websiteFits on price and contextFits on price and context
Write and review code with tool callsFits on price and contextFits on price and context
Tasks it fits6 of 66 of 6

Checks of our data for six common tasks: a published price, a context window and answer length for the task and, where the task needs them, strict JSON output, page images and tool calls; prices, cost a month and terms for each model on its model card. Compare models for your own task in the model picker.

Price changes in the last 30 days

  • The listed price of Mistral Large 4 changed on on OpenRouter: input $0.68 → $1.36, output $2.09 → $4.18 per 1M tokens.
  • The listed price of Mistral Large 4 changed on on OpenRouter: input $1.36 → $0.70, output $4.18 → $2.10 per 1M tokens.
  • The listed price of Mistral Large 4 changed on on the provider's price list: input $1.36 → $0.70, output $4.18 → $2.10 per 1M tokens.

All model comparisons · How we check prices

Explore the numbers

Calculated from published rates

The same work. Two bills.

Input and visible output are priced separately for the same task.

$0.092L2 task differenceMistral Small 4 less
$1,342.47L2 monthly differenceAssumption: 20 tasks/hour × 730 hours
Mistral Large 4$0.118Mistral Small 4$0.026$0.00$0.05$0.10$0.15Mistral Large 4$0.118Mistral Small 4$0.026$0.00$0.05$0.10$0.15
Fresh inputVisible output

159,000 input + 3,000 visible output tokens per task. No cache discount; hidden reasoning excluded.

Mistral Small 4 costs $0.092 less per task.

Mistral Large 4$0.0015Mistral Small 4$0.0003$0.00$0.0005$0.001$0.0015Mistral Large 4$0.0015Mistral Small 4$0.0003$0.00$0.0005$0.001$0.0015
Fresh inputVisible output

1,500 input + 200 visible output tokens per task. No cache discount; hidden reasoning excluded.

Mistral Small 4 costs $0.0011 less per task.

Mistral Large 4$0.940Mistral Small 4$0.203$0.00$0.50$1.00Mistral Large 4$0.940Mistral Small 4$0.203$0.00$0.50$1.00
Fresh inputVisible output

1,315,500 input + 9,000 visible output tokens per task. No cache discount; hidden reasoning excluded.

Mistral Small 4 costs $0.737 less per task.

Mistral Large 4$1,716.96Mistral Small 4$374.49$0$500$1,000$1,500$2,000Mistral Large 4$1,716.96Mistral Small 4$374.49$0$500$1,000$1,500$2,000
Fresh inputVisible output

Assumption: 20 tasks/hour × 730 hours = 14,600 tasks/month.

Observed · United States

Wait and streaming speed

The same short request. Each route must have successful answers on at least two dates.

Mistral Large 4

Direct API · Mistral · 65 answers on 13 dates.

First answer token 634 ms
Observed range · dot = median0.02.04.06.08.0secondsObserved range · dot = median0.02.04.06.08.0seconds

Mistral Large 4 · 634 ms median; observed range 489 ms–7.39 s.

Output speed 91.3 tokens/s
Observed range · dot = median0100200300400tokens/sObserved range · dot = median0100200300400tokens/s

Mistral Large 4 · 91.3 tokens/s median; observed range 12.9 tokens/s–212 tokens/s.

Observed · independent medians

More output than the answer shows.

Visible answer64Hidden reasoning0020406080Visible answer64Hidden reasoning0020406080

Calculated median cost per request: $0.000376.

Visible outputHidden reasoning

Visible output and hidden reasoning are independent medians and are not added together. Request cost is the median of individually priced requests. Hidden reasoning is billed at the output rate.

Published context window: 1,048,576 tokens.

Mistral Small 4

Collecting

No route for this model has passed the publication threshold.

Published context window: 262,144 tokens.

Unknown

No comparable published API documentation for this pair.

Sources: this page; published model prices (data (JSON)); task assumptions (data (JSON)). Published prices · Observed responses · Calculated scenarios · Assumptions stated with each chart.