ENPT

Verdin-4B

Last updated:

Open—$0/M input

Quadrant scores

See the full quadrant

Scored with the public quadrant rubric (maturity × capability; bubble = adoption). Revise as evidence lands.

  • Maturity4.7/10

    Qwen3.5-4B plus a merged LoRA and pointer head; weights CC-BY-NC-4.0 (non-commercial), Verdin engine Apache-2.0 for CUDA, CPU and Apple silicon (MLX), serving TypeSafe-style POST /v1/systemone. Independent author, no SLA.

  • Capability6.2/10

    Decision Index 0.4.1 (22 Oct) Full 46.74 (#91), below Jev 1.13 (59.22). Choice, noul and score in one shared prefill, state read bidirectionally. No ECE and no latency figures published.

  • Adoption1/100

    No downloads counted yet and two likes (20 Oct). No integrations or production users found.

Vendor claims

publicado (OK do Fabricio 20/10 BRT). Scores are calculated from scratch.

What it is

Independent 4B decision model with its own engine for CUDA, CPU and MLX and a /v1/systemone server; non-commercial weights. Decision Index 0.4 Full 49.31 (#71). Source: huggingface.co/victorbennett/Verdin-4B.

Board result

Board Row Result
Decision Index 0.4 (20 Oct 2026) Verdin-4B Full 49.31 (#71); Jev 1.13 61.26 (#34), Jev 1.14 62.22 (#28)

Not checked against JevBench or Artificial Analysis for this page; the boards are never combined.

Vendor claims (UnverifiedClaim)

  • On an M5 Max, MLX picks the same option as the CUDA engine on 99.74% of 8,129 evaluation questions (source)

Scores (from scratch, 20 Oct 2026)

  • Capability 6.2 = types × 0.30 + calibration × 0.25 + latency × 0.25 + accuracy × 0.20 = 10 × 0.30 + 4 × 0.25 + 4 × 0.25 + 6 × 0.20 = 6.20. Accuracy: 8 × 49.31 ÷ 61.26 = 6.44 → 6.
  • Maturity 4.7 = availability × 0.30 + docs × 0.25 + integrations × 0.25 + support × 0.20 = 5 × 0.30 + 7 × 0.25 + 4 × 0.25 + 2 × 0.20 = 4.65 → 4.7.
  • Adoption 1 = engagement × 0.35 + ecosystem × 0.35 + production × 0.30 = 1 × 0.35 + 1 × 0.35 + 0 × 0.30 = 0.70 → 1.

Board audit (22 Oct 2026, DI 0.4.1)

publicado (OK do Fabricio 22/10 BRT).

Decision Index 0.4.1 (22 Oct; reference Jev 1.13 Full 59.22, was 61.26 on 0.4) lists this page’s row at Full 46.74 (#91). Accuracy rule: 8 × 46.74 ÷ 59.22 = 6.31 → 6 (was 6). Capability unchanged. Maturity and adoption unchanged.