Mixtral 8x22B Instruct
Context 65K tokens
Providers
A price exists, but your data filter hides it.
Mistral is not on OpenRouter's list of providers that neither train on nor retain prompts, and such providers are left out while “Trains or keeps your data” is off.
Composite
1 of 7 inputs6 radar axes: DesignArena's two boards share one1.7
Thin data · 1/7Based on only 1 of 7 Composite inputs — treat this rank as uncertainAA Coding v4.x —Coding Agent v1.4 —AA Intelligence (est.) 2.3AA Agentic —Epoch ECI —Software ECI —DesignArena —/—
Radar: percentile among all models measured on each input; a gap means not measured.
Benchmark sheet
6 of 319 registered benchmark versions · bars show the percentile among all models measured on each benchmark.
Knowledge
Humanity's Last Exam (AA text-only)134.0%
Humanity's Last Exam (AA text-only) published 2026-09-10 → · Published board — Tests expert-level knowledge on AA's text-only Humanity's Last Exam subset.
UGI NatInt5632.3
UGI NatInt published 2026-09-10 → · Published board — A private-question evaluation of general knowledge and reasoning across textbook, popular-culture and world-model topics.
Science
GPQA Diamond (AA)1033.2%
GPQA Diamond (AA) published 2026-09-10 → · Published board — Tests graduate-level biology, physics and chemistry knowledge on the Diamond subset.
Uncensored
UGI Willingness623.50
UGI Willingness published 2026-09-10 → · Published board — A private-question evaluation of how far a model follows challenging instructions before refusing or deviating.
UGI5838.9
UGI published 2026-09-10 → · Published board — A private-question suite combining sensitive-topic knowledge with willingness to follow controversial instructions.
Writing
UGI Writing2727.4
UGI Writing published 2026-09-10 → · Published board — A maintainer-owned writing evaluation considering intelligence, style, repetition and length adherence, informed by human preferences.
Missing coverage · 313 benchmark versions
No result does not mean a zero, or that the model was never tested. Collection failures and disputed versions retain their distinct status.
Token offers by platform · 0 offers (Adjusted $/task)
Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.
0 offers within the active global filters, 3 more outside them; “price n/a” means the provider serves the model but publishes no direct token price we could collect — no other channel's price is shown in its place.
All 3 offers of this model are outside the active global filters.