H2O-Lightning-4B

Released
2026-10-02
Last refreshed
2026-10-12
Status
Researched today
Open sourceCommercial use: permittedMultimodalAgentsVisionJSON / Tool useAgent

H2O-Lightning-4B is available now for agents, vision, and json / tool use with open-source; evaluate it while provider pricing coverage matures.

H2O.ai releases · 1 in the last 12 monthsChangelog →
Specifications
Released
2026-10-02
Context
40K
Parameters
4B
Openness
Open source
License
Apache 2.0OSI-approvedCommercial use: permitted
Weights
Available
Code
Unknown
Training
Fine-tuned
Created by
H2O.ai

H2O.ai is an open-source machine learning company. In October 2026 it released H2O-Lightning-4B, an Apache-2.0 decision

Website
Pricing

No tracked provider token pricing is available yet.

Links
Website HuggingFace

About

H2O-Lightning-4B is H2O.ai's open-weight decision model, a fine-tune of Qwen3.5-4B released on Hugging Face under Apache 2.0 (HF repo created 2026-10-02). It answers typed choice, noul (yes/no probability) and score questions about text records and, in current revisions (card documents v1.2.x image path), about up to four images, returning a probability for every option in one forward pass with no generated explanation. It runs on unmodified vLLM 0.30.0 behind the shipped `h2o_lightning_shim.py` (`POST /v1/systemone`). Context up to 40,960 tokens. Not a chat model.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Capabilities

VisionMultimodalStructured Outputs

Benchmark peer barsfor Agents

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

API versions

H2O-Lightning-4Bh2o-lightning-4bh2oai/h2o-lightning-4bH2O-Lightning-4B v1.1H2O-Lightning-4B v1.2.3