Skip to content
ModelCap

Model decision surface

Compare AI models

Start with LFM2 8B A1B and Qwen2.5 Coder 32B Instruct, 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

LFM2 8B A1B vs Qwen2.5 Coder 32B Instruct

Live dataset updated 9/12/2026, 1:09:54 PM UTC

Open 1200×630 evidence receipt
Factual comparison of LFM2 8B A1B and Qwen2.5 Coder 32B Instruct
Field
LFM2 8B A1B

LiquidAI

ModelCap position#170#173
Index score18.616.3
EvidenceEstimatedglobal-corpus-prior over 188 held-out anchors (19%); launch card against 3 resolved peers on 10 rows (81%); exceeds every named peer on 8 of 10 rows; optimism haircut 9.5 from cross-lab-probe; shrunk 8.1 up toward the measured corpusMeasured2 public benchmark observations across 2 boards
Input / 1MUnavailable$0.66
Output / 1MUnavailable$1.00
Pricing statusunavailablefresh
Context33K
Providers01
Weight accessRestricted licenseOpen weights

Decision facts

  • LFM2 8B A1B is #170; Qwen2.5 Coder 32B Instruct is #173 on the same current language board.
  • Index scores are 18.6 for LFM2 8B A1B and 16.3 for Qwen2.5 Coder 32B Instruct. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
  • Evidence differs: LFM2 8B A1B is Estimated; Qwen2.5 Coder 32B Instruct is Measured.
  • Weight access differs: LFM2 8B A1B is restricted; Qwen2.5 Coder 32B Instruct is open.
  • ModelCap currently lists 0 providers for LFM2 8B A1B and 1 for Qwen2.5 Coder 32B Instruct.

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.