Issues

• [bench request]: DecisionNode Edge 0.9.2 and Edge Mini 0.9.2 (open weights, /v1/systemone server)

  fstandhartinger/jevbench#231 · Filter by author bynnit opened on Oct 23, 2026
  1

• Benchmark proposal: DataJevBench — a data-processing decision evaluation for JevBench

  fstandhartinger/jevbench#230 · Filter by author helloml0326 opened on Oct 22, 2026
  1

• Benchmark proposal: HarnessJevBench — a harness runtime control evaluation for JevBench

  fstandhartinger/jevbench#229 · Filter by author helloml0326 opened on Oct 22, 2026
  1

• [bench request]: Mica v0.5 4B (open weights, LoRA fine-tune of Qwen3.5-4B, /v1/systemone server)

  fstandhartinger/jevbench#228 · Filter by author akivet opened on Oct 22, 2026
  1

• [bench request]: baselabs/mammoth-1 (27B, open weights, public)

  fstandhartinger/jevbench#227 · Filter by author team-sqwish opened on Oct 22, 2026
  1

• [bench request]: Yev-Decision-1 (open weights, 4B, runs locally)

  fstandhartinger/jevbench#225 · Filter by author CodeHalwell opened on Oct 21, 2026
  1

• System submission: Selena OS System One Decision Core (selena-v1-system-one)

  fstandhartinger/jevbench#224 · Filter by author JanuszSikora-Selena-OS opened on Oct 20, 2026
  3

• Re-measurement request: Seb-9B (ref C566D021) — card fixed (base model, 32k context)

  fstandhartinger/jevbench#222 · Filter by author IronBCC opened on Oct 19, 2026
  1

• Dataset proposal: financial market event prediction

  fstandhartinger/jevbench#221 · Filter by author xiangchen-Z opened on Oct 19, 2026

• Jebadiah 9B v2.1 and 27B v2.1 for JevBench

  fstandhartinger/jevbench#220 · Filter by author webdevtodayjason opened on Oct 16, 2026
  1

• [bench request]: Chon-1.0-12B (open weights, self-hosted)

  fstandhartinger/jevbench#219 · Filter by author xoai opened on Oct 15, 2026
  1

• [bench request]: Amos 1 large (27B, Qwen3.8-27B, Apache-2.0): image input, native option probabilities, System One wire format, open weights + hosted API

  fstandhartinger/jevbench#218 · Filter by author zaidmukaddam opened on Oct 15, 2026
  1

• Evaluation request: Respan Span-02 - Open Weights, Self-hosted

  fstandhartinger/jevbench#215 · Filter by author lihe07 opened on Oct 13, 2026
  4

• latency measurement fix

  fstandhartinger/jevbench#212 · Filter by author pseudotensor opened on Oct 11, 2026
  1

• Benchmark request: Krino-2B (Kev post-train on Qwen3.5-2B-Base, Apache-2.0)

  fstandhartinger/jevbench#211 · Filter by author guru0381 opened on Oct 10, 2026
  1

• Brazilian Portuguese typed-decision track

  fstandhartinger/jevbench#202 · Filter by author antonio-mello-ai opened on Oct 10, 2026
  1

• [bench request]: Strands Decider v21 pruned 1.4B (unofficial; Qwen3.5-2B-Base, 15 of 24 layers, native typed probabilities, Apache-2.0)

  fstandhartinger/jevbench#201 · Filter by author alxnahas opened on Oct 9, 2026
  1

• [bench request]: jul fast (usejul/jul-decision-minicpm5-2b; MiniCPM5-2B fine-tune + pointer head, native option probabilities, TypeSafe wire format via jul serve, Apache-2.0)

  fstandhartinger/jevbench#198 · Filter by author jlestel opened on Oct 7, 2026
  1

• [bench request]: mc-one-draft 12b (frozen google/gemma-4-12B-it, one-token letter readout, TypeSafe wire format)

  fstandhartinger/jevbench#196 · Filter by author kienbui1995 opened on Oct 7, 2026
  3

• [bench request]: Add Messier One (Qwen3.5-4B fine-tune, native option probabilities, TypeSafe wire format)

  fstandhartinger/jevbench#194 · Filter by author agentmessier-ai opened on Oct 6, 2026
  4

• Image JevBench: image-input limits of Perplexity's Decisions API and of XOR

  fstandhartinger/jevbench#192 · Filter by author murdore opened on Oct 4, 2026
  1

• Model evaluation request: decision-stack-0.8b-v0.1 (runnable offline bundle)

  fstandhartinger/jevbench#190 · Filter by author mrmps opened on Oct 4, 2026
  2

• [bench request] Image JevBench: Image Hopper (Qwen3.5-9B, routed LoRA, calibrated one-pass option probabilities)

  fstandhartinger/jevbench#189 · Filter by author ArchieIndian opened on Oct 4, 2026
  3

• [bench request]: Quyet 1.0 — five open-weight decision models (Large 31B, Medium 4B, Small 300M, Small-EN 150M, Tiny 180M; native option probabilities, in-process adapter, Apache-2.0)

  fstandhartinger/jevbench#188 · Filter by author ncchinh opened on Oct 4, 2026
  1

• Proposal: Traditional Chinese (zh-TW) evaluation slice, with Plumb-4B results on Ollama

  fstandhartinger/jevbench#187 · Filter by author wuchengwei1996 opened on Oct 4, 2026
  1