JevHunt

club-170hx: Community-tested recipes, diagnostics, and benchmarks

PixelML/club-170hx

VerifiedResearchDocumented

club-170hx: Community-tested recipes, diagnostics, and benchmarks for CMP 170HX AI workloads. Explore the Jev references, Jupyter Notebook source code and similar projects.

Repository description: Community-tested recipes, diagnostics, and benchmarks for CMP 170HX AI workloads

Open GitHub ↗ Read evidence ↗ Alternatives Report a problem

Apache-2.0 · Jupyter Notebook · updated 2026-09-27 · 62 ★ · Evidence checked Sep 29, 2026

Get started

Clone the repository and follow its README for install and usage. JevHunt has not run this project.

git clone https://github.com/PixelML/club-170hx.git

Open the README ↗

Project facts

Relationship to Jev
Research
Evidence
Documented
Language
Jupyter Notebook
License
Apache-2.0
Origin
Original repository
Repository status
Not archived
Created
2026-08-30
Last code update
2026-09-27
GitHub stars
62
Evidence checked
Sep 29, 2026
Metadata checked
Sep 29, 2026
Check status
Up to date

Stars measure the whole repository, including work unrelated to Jev.

Evidence and scope

Documented records the linked documentation or source. JevHunt has not independently run or benchmarked this project.

| 2026-09-20 | Qwen3.8-27B W4A16 GPTQ as a **no-train Jev-compatible endpoint** (`/v1/systemone`), 1x CMP 170HX, vLLM sm80 | **Classification works on this card and survived its first production workload; the calibration is the project. No-train method: stock W4A16 checkpoint served read-only, and the only fitted quantity (a temperature) failed leave-one-out. Production (2026-09-21): an LLM-as-judge annotation pilot, 19,045 annotations in 4.5 h, 0 failures, 5 permutation reads/annotation, 1,952

Evidence commit: 17fe827c209d3c20a4a1b065bed1ddac9e504283

Discovered through: github-search.

How we review →

Suggest a correction · Claim this listing · Submit a project