Jev directory
CommunityJev Models & Reimplementations
Strands-Decider E2B Gemma4 v1 2610
A Strands decider v1 checkpoint: a LoRA adapter and readout head on google/gemma-4-E2B-it that answers typed decisions with calibrated confidence; its model-index reports 0.7792 JevBench public accuracy (180/231) and code-bug evals from 0.281 to 0.978.
- Reading time
- 2 min
- Published by
- Community
- Author
- Strands Agents
- Last verified
- Oct 2026
- Capabilities
- Open sourceSelf-hostedRuns locally
Highlights
- LoRA adapter plus a small readout head on google/gemma-4-E2B-it; model-index reports 0.7792 JevBench public accuracy (180/231).
- Reports Musique 0.913, ContractNLI 0.856, boardgame 0.793, hotpotqa 0.734 and held-out short tasks 0.652 (n=6000).
- Code evals: commit_message 0.978 and docstring_match 0.966, but correct_solution 0.281 and bug_wrong_binary_operator 0.674, each on 89 items.
- Trained on public datasets including HelpSteer2, commitpackft, cubert_ETHPy150Open and juliet_test_suite_c_1_3; Apache-2.0 adapter.
Watch out
Card reviewed 2026-10-17; the repository has no prose card, so this entry rests on the front matter and self-reported model-index results rather than an independent evaluation. The adapter declares Apache-2.0 while the gemma-4-E2B-it base keeps Google's Gemma terms.
Questions about this entry
Short answers drawn from the linked project or page.
A v1 checkpoint of the Strands decider family: a LoRA adapter plus a small readout head on google/gemma-4-E2B-it that scores the options of typed questions with calibrated confidence instead of generating text.
JevBench public at 4096-token serving 0.7792 (180/231) and a list of publisher eval configs: held-out short tasks 0.652 (n=6000), boardgame 0.793, ContractNLI 0.856, hotpotqa 0.734 and Musique 0.913.
Mixed: commit_message 0.978, docstring_match 0.966 and cwe_choice 0.899, but correct_solution 0.281 and bug_wrong_binary_operator 0.674, each measured on 89 items, so weak spots show up clearly.
It is an adapter on google/gemma-4-E2B-it, declared Apache-2.0 in the repository front matter, while the Gemma base keeps Google's own Gemma terms, so check both before commercial use.
Like the rest of the family it answers typed decision questions (choice, noul, score) with probabilities rather than text, following the System One style the Strands decider line targets; this card itself carries no API or serving notes.
Same recipe on a smaller base: the E4B uses google/gemma-4-E4B-it and reports 0.8095 JevBench (187/231) and 0.684 short-task accuracy, against this model's 0.7792 and 0.652.
No. The repository holds front matter and model-index results only, with no prose usage notes or commands, so serving specifics would have to come from the wider Strands decider codebase, which this card does not link.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Standard One SH (StandardThinking)
LinkedIn: What multimodal decision models actually change, and what they do not.
LinkedInJevDirectory.org
Hacker News: Decision Index 0.2.1 scores open reproductions of Jev on a frozen suite.
jackbravo4 pointsDecision Index 0.2.1 – 70 open reproductions of TypeSafe Jev's Decision ModelRead the thread on Hacker NewsHacker News: Pointer to a DiffusionGemma plus vLLM tutorial for a Jev-shaped classifier.
1 pointHow to build a Jev-style classifier with DiffusionGemma and vLLMRead the thread on Hacker NewsMore Jev resources like this
A Strands decider v1 checkpoint: a LoRA adapter and readout head on google/gemma-4-E4B-it that answers typed decisions with calibrated confidence; its model-index reports 0.8095 JevBench public accuracy (187/231) and mixed code-bug evaluation scores.
Jev Models & Reimplementations#community#open-models#local
Community
A pointer-network decision model from Keyush Nisar that reads one agent state document plus typed questions and returns calibrated tool-routing probabilities in a single forward pass, reporting 89.00% zero-shot tool routing.
Jev Models & Reimplementations#community#open-models#local
Community
A Laya fine-tune for coding decisions: roughly 9k rows from SWE-Gym teach keep/drop file relevance, priority scoring and injection detection, reporting 0.833 accuracy on 1,150 held-out rows at 126 ms p50.
Jev Models & Reimplementations#community#open-models#local
Community
The first v1 checkpoint of the Strands decider line from the Strands Agents team: a LoRA adapter and readout head on Qwen3.5-2B-Base that answers typed decisions with calibrated confidence, reporting 0.7792 JevBench public accuracy (180/231).
Jev Models & Reimplementations#community#open-models#local
Community
Jared Palmer's 27B Kev decision model: a full-weight fine-tune of Qwen3.8-27B averaged with v1, plus a pointer head that returns calibrated probabilities for typed choice, score and noul questions in one forward pass. Card-reported transfer-v4 accuracy is 88.87%.
Jev Models & Reimplementations#community#open-models#local
Community
Mapika's 4B decider: a Qwen3.5-4B-Base fine-tune that returns probabilities over supplied options for typed questions in one forward pass, with no decoding. Version 2.1 adds a rank-64 LoRA stage and reports a JevBench public hard tier of 0.649.
Jev Models & Reimplementations#community#open-models#local
Community
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.