jaredpalmer/Kev 0.8B

• License: Apache-2.0
• Release date: 2026-09-24 (Kev 1.0 checkpoint, as documented in the original model card)
• Modalities: Input: text; Output: typed decisions (choice, score, noul)
• Original model: jaredpalmer/kev-0.8b

Description

Kev reads a document (the state) and typed questions, returning calibrated probabilities over the supplied options in one forward pass without generating text. This model uses a LoRA adapter and a pointer head on Qwen3.5-0.8B-Base. It targets memory- or cost-constrained applications; the upstream model card reports lower out-of-domain accuracy than Kev-4B.

Run

self serve kev:0.8b

Here is a link to the API docs: Decision API.

Example Request

With the server running, open another terminal and send a request:

curl http://localhost:8080/v1/systemone \
	-H 'Content-Type: application/json' \
	-d '{
		"state": "I was charged twice for the same order.",
		"questions": {
			"department": {
				"type": "choice",
				"instructions": "Which department should handle this support ticket?",
				"criteria": {
					"billing": "Payments and refunds",
					"technical": "Technical problems"
				}
			}
		}
	}'

The server returns a JSON decision for the department question.

Engine Parameters

Served by the upstream laya daemon through the ggmlc-laya adapter.

Parameter	Default	What it controls
threads	4	Number of CPU workers; accepts integers from 1 to 1024.
cuda_graph	true for auto / CUDA devices; otherwise false	Capture a CUDA graph for the live request shape.

Override startup settings with --set <parameter>=<value>.

Quantizations and Files

Default variant: q4. Sizes and SHA-256 hashes are pinned in the model registry.

q4 (default)

File	Role	Size	SHA-256	Origin
kev_0.8b_ud_q4_k_m.gguf	model	914.3 MiB (958685280 bytes)	410d9c071a72435d5579b88af98b53fe520260ae61e4154c234e6a28a4ae9623	mys/kev-0.8b-GGUF

Total download: 914.3 MiB.

q8

File	Role	Size	SHA-256	Origin
kev_0.8b_q8_0.gguf	model	789.5 MiB (827809856 bytes)	12f77506dbfaa3c1d12f7e3f8da04feceda276f2d5aa322cd0858a8ce11e6c6c	mys/kev-0.8b-GGUF

Total download: 789.5 MiB.

fp16

File	Role	Size	SHA-256	Origin
kev_0.8b_f16.gguf	model	1.4 GiB (1533202496 bytes)	6b15f6bd031e1f0f3b751692aaff75b99d1dbb5362f4cda207b3d4d26e5e11eb	mys/kev-0.8b-GGUF

Total download: 1.4 GiB.

Notes

• Text only; up to 16 options per question.
• Context and option limits are compiled into the GGUF and cannot be changed at runtime.
• Download size is not the total runtime memory requirement.
• Origin links use main because these registry entries pin hashes, not repository revisions.

Quality benchmarks

Decision Index: #133 (13.66)

jaredpalmer/kev-0.8b

Published result for the named upstream model and evaluated configuration; checkpoint revision and local quantization equivalence are not guaranteed.

Evaluated: kev-0.8b-r15

Edition: Decision Index 0.3.2

Checked 2026-10-13

Hugging Face model-card evaluations: Publisher-reported

jaredpalmer/kev-0.8b

Publisher-reported upstream checkpoint; not an independent rank or local quantization

Checked 2026-10-13

Dataset	Metric	Score
transfer-v4 test: six never-trained public sources and held-out policy structures (656 questions)	accuracy	0.70
transfer-v4 test: six never-trained public sources and held-out policy structures (656 questions)	brier_score	0.40
breadth-v1 test: 14 held-out public datasets (3,089 questions)	accuracy	0.59
tasksource-heldout-v1 development: 17 of 24 held-out task families (1,993 questions)	accuracy	0.52
documents-v1 test (936 questions on CFPB complaint narratives)	accuracy	0.85
documents-v1 test (936 questions on CFPB complaint narratives)	brier_score	0.24
hard-v1 test (1,088 questions; programmatic labels, held-out templates)	accuracy	0.67
hard-v1 test (1,088 questions; programmatic labels, held-out templates)	brier_score	0.46
devtools-v1 test (1,071 questions; six public developer-tooling sources)	accuracy	0.64
devtools-v1 test (1,071 questions; six public developer-tooling sources)	brier_score	0.44

Benchmarks

No submitted benchmark reports yet.