Add a choice 2/4

Showing 8 of 596. Type to narrow. Up to 4 choices side by side.

WHAT THE EVIDENCE SAYS

Mistral Large 4 has the lowest estimated workload cost: $67.90 per month.

Editorial coverage is incomplete. Unscored models are not assigned invented scores.

Latency and reliability are not ranked without comparable measured conditions.

Workload assumptions 100 workloads / month

Text-token estimates only. Cache creation, cache storage, tools, taxes, and media are excluded. Without a documented cache-read rate, input is charged at the full listed rate.

Marks a row where choices differ. Unknown means unknown, not zero.
Decision criteriaMistral Large 4Mistral · API modelGrok 4.5xAI · API model
Overview
TypeAPI modelAPI model
ProviderMistralxAI
Best suited forAccepts text, images and produces text. Supports tool calling and reasoning controls.Teams that need a documented SWE-bench Pro number from xAI's line rather than 4.6's unpublished claims
Fit and cost
Saved editorial task fitNot scored74.9 / 100 · editorial
Monthly workload estimate$67.90$201.00
Price basisverified · checked Nov 14, 2026verified · checked Nov 14, 2026
Capability and access
Context window1,048,576 tokens500,000 tokens
Maximum output262,144 tokensn.a. on fetched pages
License / weightsNot documentedProprietary
AccessOpenRouter API; availability and price can vary by routed provider.xAI API and console, Grok Build (was the default model), Cursor on all plans
Evidence and limits
Important limitationsNo normalized editorial score or independently checked benchmark is assigned to this catalogue addition.Superseded by Grok 4.6, which scores higher on both preference boards at the same price · AA publishes two conflicting Intelligence Index values (45 vs 56) · 500K context with an undisclosed >200K surcharge · Cached-input rate is an estimate, not an xAI-published price
Source links2 linked sources6 linked sources
Benchmark observations0 observations; inspect conditions6 observations; inspect conditions
Inspect sourcesFull profile & evidence Full profile & evidence
Editorial assessmentVendor / source reportMeasured observationCost estimate

These evidence types are not interchangeable. Benchmark conditions and pricing sources are available in each profile.