Skip to content
ModelCap

Model decision surface

Compare AI models

Start with Mistral Small 4 and Step 3.7 Flash, or choose any two current ranked language models. Compare capability evidence, price, context, provider availability, and weight access without pretending one field decides every use case.

Current public data

Mistral Small 4 vs Step 3.7 Flash

Live dataset updated 9/22/2026, 10:37:09 PM UTC

Open 1200×630 evidence receipt
Factual comparison of Mistral Small 4 and Step 3.7 Flash
Field
Mistral Small 4

Mistral AI

ModelCap position#97#96
Index score54.154.2
EvidenceMeasured1 public benchmark observation across 1 boardEstimated3 reported rows against measured corpus ladders (20%); global-corpus-prior over 209 held-out anchors (17%); measured predecessor stepfun/step-3.5-flash less the succession penalty (63%); shrunk 0.6 up toward the measured corpus
Input / 1M$0.15$0.20
Output / 1M$0.60$1.15
Pricing statusfreshfresh
Context262K262K
Providers13
Weight accessOpen weightsOpen weights

Decision facts

  • Mistral Small 4 is #97; Step 3.7 Flash is #96 on the same current language board.
  • Index scores are 54.1 for Mistral Small 4 and 54.2 for Step 3.7 Flash. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
  • Evidence differs: Mistral Small 4 is Measured; Step 3.7 Flash is Estimated.
  • Listed output price per 1M tokens is $0.60 for Mistral Small 4 and $1.15 for Step 3.7 Flash. For 1,000 requests with 2,000 input and 500 output tokens each (2M input + 0.5M output), the listed-rate estimate is $0.60 for Mistral Small 4 versus $0.97 for Step 3.7 Flash. Mistral Small 4 costs 38.5% less in this scenario. This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the selected endpoint before budgeting.
  • Published context is 262,144 tokens for Mistral Small 4 and 262,144 for Step 3.7 Flash.
  • ModelCap currently lists 1 providers for Mistral Small 4 and 3 for Step 3.7 Flash.

These are separate published fields, not a synthetic winner. ModelCap does not collapse price, access, context, and capability evidence into a hidden recommendation score.

Each comparison page is a permanent, shareable URL with the same live figures as this tool: ModelCap Index position, API pricing, context window, provider count, weight access and every shared benchmark board.