h2o-lightning-4b
h2o-lightning-4b by H2O.ai is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text.
Pricing
No hosted price is listed. The weights are open: the cost is your own compute.
Specifications
- Provider
- H2O.ai
- Parameters
- 5B
- Released
- Oct 2026
- Open weights
- Yes
- Question types
- Choice (one of N options), Score (ordered rubric), Yes / no
- Mechanism
- Decoder LM with a decision readout
- Backbone
- Qwen/Qwen3.5-4B
- Jev-compatible request
- Yes
- Runs on CPU
- No
- License
- apache-2.0
- Hugging Face
- h2oai/h2o-lightning-4b
Benchmarks
Compare h2o-lightning-4b with…
FAQ
How much does h2o-lightning-4b cost?
h2o-lightning-4b has no hosted price listed; its weights are open, so the cost is the compute you run it on.
What questions can h2o-lightning-4b answer?
h2o-lightning-4b answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.
Is h2o-lightning-4b well calibrated?
No calibration measurement for h2o-lightning-4b is recorded yet.
Is h2o-lightning-4b open weight, and can it run on a CPU?
Yes, h2o-lightning-4b has open weights. It needs a GPU to run at useful speed.
Data as of 2026-10-14. Source: Metatext decision-model catalogue, 2026-10-14 (Benchmark Heaven JevBench, Hugging Face). Pricing is per 1M tokens (USD); confirm with the provider before production use.