JevK5
JevK5 is an independent, Apache-2.0 open-source alternative to TypeSafe's Jev for typed decisions. It reads a state and a yes/no (noul), choice, or score question and returns a probability for every option in one forward pass, with zero generated tokens. The open weights can be self-hosted using the JevK5 runtime, which serves a TypeSafe-style /v1/systemone endpoint. This is not Jev's model or architecture and is not...
Params
4.2 B
Context
262,144
Downloads 30d
138 K
Likes
27
Download history
tracking started — chart appears after 7 days of snapshots (2 recorded)138 K downloads in the last 30 days
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | bf16 | 8.4 GB | 10.4 GB | ✅ Runs comfortably |
| model.safetensors (bf16, full) | bf16 + 256K ctx | 8.4 GB | 29.9 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ curl -s https://aimodelscomparison.com/api/v1/models/jevk5
{
"hf_id": "alibiserikbay/JevK5",
"params_b": 4.21,
"context_length": 262144,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 138399,
"vram_estimates": [
{ "quant": "bf16", "gb": 10.4 }
],
"updated_at": "2026-11-09T01:00:11Z"
}
Specifications
- Architecture
- Qwen3_5ForCausalLM
- Parameters
- 4.2 B
- Tensor type
- BF16
- Context length
- 262,144
- Vocabulary
- 248,320
- Layers / heads
- 32 / 16
- Licence
- apache-2.0
- First seen on the Hub
- 2026-09-22
- Base model
- Qwen3.5-4B
- Training datasets
- undisclosed
- Added to our catalog
- 2026-11-09
Compare with any text-generation model