Mistral: Mixtral 8x22B Instruct

Public pricingIntelligence 70/100Medium memoryใƒ„ใƒผใƒซๅˆฉ็”จ

Mistral: Mixtral 8x22B Instruct ใฏ ใƒ†ใ‚ญใ‚นใƒˆ ใƒขใƒ‡ใƒซใงใ€ใ‚ณใƒผใƒ‡ใ‚ฃใƒณใ‚ฐใ€ใƒ‡ใƒใƒƒใ‚ฐใ€ๆŠ€่ก“ไฝœๆฅญ ใซๅ‘ใ„ใฆใ„ใพใ™ใ€‚้ซ˜ใ„ใ‚ณใƒผใƒ‡ใ‚ฃใƒณใ‚ฐๆ€ง่ƒฝใ€ไฝŽ้…ๅปถใงๅŠน็އ็š„ใชๆŽจ่ซ–ใ€66K tokensใฎใ‚ณใƒณใƒ†ใ‚ญใ‚นใƒˆใ€ใƒใƒฉใƒณใ‚นๅž‹ใ‚ณใ‚นใƒˆใฎ็‰นๆ€งใ‚’็ต„ใฟๅˆใ‚ใ›ใ€coding, debugging, and technical writing ใงๅฎ‰ๅฎšใ—ใŸๅ‹•ไฝœใ‚’ๆ”ฏใˆใพใ™ใ€‚้…ๅปถใƒปใ‚ณใ‚นใƒˆใƒปใ‚นใƒซใƒผใƒ—ใƒƒใƒˆ ใŒ้‡่ฆใชๅ ด้ขใซๅˆใฃใฆใŠใ‚Šใ€ๅฎ‰ๅฎšใ—ใŸๅ‡บๅŠ›ใ€ๆŸ”่ปŸใชๅฐŽๅ…ฅใ€ๆ‹กๅผตๆ€งใ‚’ๆฑ‚ใ‚ใ‚‹ใƒใƒผใƒ ใซๅฎŸ็”จ็š„ใงใ™ใ€‚ ๅฎ‰ๅฎšใ—ใŸๅฟœ็ญ”ใ€้•ทใ„ๆ–‡่„ˆๅ‡ฆ็†ใ€ใใ—ใฆ่ฉฆไฝœใ‹ใ‚‰ๆœฌ็•ชใพใงๅบƒใไฝฟใˆใ‚‹ๆŸ”่ปŸใ•ใŒๅฟ…่ฆใชๅ ด้ขใงๅฝน็ซ‹ใกใพใ™ใ€‚ ๅฎ‰ๅฎšใ—ใŸๅฟœ็ญ”ใ€้•ทใ„ๆ–‡่„ˆๅ‡ฆ็†ใ€ใใ—ใฆ่ฉฆไฝœใ‹ใ‚‰ๆœฌ็•ชใพใงๅบƒใไฝฟใˆใ‚‹ๆŸ”่ปŸใ•ใŒๅฟ…่ฆใชๅ ด้ขใงๅฝน็ซ‹ใกใพใ™ใ€‚ ๅฎ‰ๅฎšใ—ใŸๅฟœ็ญ”ใ€้•ทใ„ๆ–‡่„ˆๅ‡ฆ็†ใ€ใใ—ใฆ่ฉฆไฝœใ‹ใ‚‰ๆœฌ็•ชใพใงๅบƒใไฝฟใˆใ‚‹ๆŸ”่ปŸใ•ใŒๅฟ…่ฆใชๅ ด้ขใงๅฝน็ซ‹ใกใพใ™ใ€‚

Input

$2.00/1M

Output

$6.00/1M

Cached

$0.20/1M

Batch

$1.00/1M

Calculate your Mixtral 8x22B Instruct bill.

Set your workload โ€” see cost at your exact volume.

What would Mixtral 8x22B Instruct cost you?

Adjust the workload to see your monthly bill.

1,00010,00050,000250,0001M10M

Technical specifications

Mixtral 8x22B Instruct at a glance.

Memory

65,536

tokens

Max reply

4,096

tokens

Memory tier

Medium

a long report or a codebase file

Tokenizer

mistral

Released

Apr 2024

Training cutoff

Jan 2024

Availability

Public pricing

Status

active

Benchmarks

Quality benchmarks

Independent evaluations from public leaderboards. Higher is better.

  • mmlu

    77.8
  • humaneval

    45.1
  • aa_intelligence_index

    10
  • ifeval

    71.84
  • bbh

    44.11
  • mmlu_pro

    38.7

What it can do

Capabilities & limits.

  • Understands images
  • Deep step-by-step thinking
  • Uses tools / calls functions
  • Strict JSON output
  • Streams replies
  • Fine-tunable on your data

When to pick Mixtral 8x22B Instruct

  • Agentic workflows that call tools or APIs.

When to look elsewhere

  • Your workload involves images โ€” pick a vision-capable model instead.

FAQ

Mixtral 8x22B Instruct โ€” the questions we see most.

Pricing, capabilities, alternatives โ€” generated from the same data that powers the calculator above.

Get instant answers from our AI agent

At a typical workload of 50,000 conversations a month with 1,500-token prompts and 800-token replies, Mixtral 8x22B Instruct costs roughly $390 per month. Input is $2.00 /1M tokens and output is $6.00 /1M tokens.
Mixtral 8x22B Instruct has a 65,536-token context window (medium memory โ€” a long report or a codebase file). That means you can fit about 12,288 words of input and history in a single call.
Beyond text generation, Mixtral 8x22B Instruct supports calling functions / tools, strict JSON output, fine-tuning on your own data. It streams replies by default.
Mixtral 8x22B Instruct was released in April 2024, with training data cut off around January 2024.
Models in a similar class include Pixtral Large 2411, Mixtral 8x7B Instruct, Mistral Large 3 2512. The "Similar models" section below this FAQ links into each.

Still unsure?

Compare Mixtral 8x22B Instruct against 100+ other models.

Open the full wizard โ€” pick a use case, set your usage, and see side-by-side monthly costs in under a minute.