hopper is an open-weight model from HopitAI, released under other. Its published files total 129.9 MB. It draws 721 downloads a month.
Hopper models make single-pass decisions: a document, a policy and a question go in, and a probability for every allowed option comes out. One forward pass per decision; no text is generated.
Model Card
Hopper models make single-pass decisions: a document, a policy and a question go in, and a probability for every allowed option comes out. One forward pass per decision; no text is generated. They are served by github.com/hopit-ai/hopper, an HTTP server that speaks the JevBench /v1/systemone wire format. ¹ The 4B adapter's training data included passages from RACE, which is released for non-commercial research only. Do not use it commercially. See Licence and research-only notice. The general-purpose line, Hopper (G), lives at HopitAI/hopper-g. Always pin the revision: HopitAI/hopper without one resolves to main, the Qwen3.5-4B adapter, which does not fit this base. Files, checksums…
Excerpt from the card by HopitAI, licensed other.
Identity and Version
- Repository
- HopitAI/hopper
- Publisher
- HopitAI
- Task
- Not stated by the source
- Modality
- Other
- Library
- peft
- Parameters
- Not stated by the source
- Languages
- en
- Revision
- 4bbab19872ae7a12a2b3aa875c4b3988ce78c051
- First published
- 2026-09-21
- Last updated
- 2026-10-03
Files and Weights
5 files, 129.9 MB in total. The weights are 1 file totalling 129.9 MB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| adapter_model.safetensors | Weights | 129.9 MB | 8c295d3a2124 |
| adapter_config.json | Configuration | 1.4 KB | — |
| hopper.json | Configuration | 403 B | — |
| README.md | Documentation | 18.2 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- other
- Access
- Open weights, no gate
- Download size
- 129.9 MB
Released by HopitAI through its official repository on Hugging Face. Read the license.
Built From
- Adapter of Qwen/Qwen3.5-4B
- Derived from Qwen/Qwen3.5-4B
- Trained on (disclosed) allenai/ai2_arc
- Trained on (disclosed) cais/mmlu
- Trained on (disclosed) clinc/clinc_oos
- Trained on (disclosed) fancyzhx/dbpedia_14
- Trained on (disclosed) google/boolq
- Trained on (disclosed) nvidia/HelpSteer2
- Trained on (disclosed) nyu-mll/multi_nli
- Trained on (disclosed) rajpurkar/squad_v2
- Trained on (disclosed) stanfordnlp/snli
- Trained on (disclosed) tals/vitaminc
- Trained on (disclosed) tau/commonsense_qa
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 129.9 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About hopper
What license is hopper released under?
other, as its publisher declares it. Read the license text before commercial use.