AI Agent StoreFrontiersMind · Download verified
Lumma-Fev 4B v2
Run ticket routing, classification or yes/no checks on your own hardware, with probabilities for the answers you define.
Released 2026-10-29 · First reviewed 2026-10-30 · Checked 2026-11-03.
We still need comparable independent tests. You can compare the confirmed features and limits below, but we cannot yet give this version an Agent Fit or Practical Value rating. Test it on records with known answers, including cases that should go to a person.
What you can use today
You download the model files and supply the hardware and software to run them. Capacity is the amount of information a request can contain; settings and available memory can limit what fits.
Exact download ID: FrontiersMind/Lumma-fev-4B-v2. Supports English, Hindi, Tamil, Kannada and Hinglish. Score questions are accepted but were not tested by the publisher. Image input requires the optional image runtime.
Apache 2.0 weights. Requires lumma-fev 0.3 or later; older versions reject this attention setup. The publisher suggests about 10 GB GPU memory for 4B. Allow extra room for your workload.
The shared information plus one question must fit in 8,192 tokens. The parent model’s larger window does not apply to this decision model.
Output: Choice and yes/no probabilities; no generated reply. Compare decision models, limits and input costs ↗
No hosted token price or measured local operating cost is established. Publisher accuracy and confidence tests remain separate from independent rankings.
Provider release and specifications ↗
Check availability and plans ↗
Test it before switching
Give each model the same examples, tools and definition of success. Record correct outcomes, response time, retries and total cost. Include the time and hardware needed to run it.
Compare caching and long-context costs ↗