BLK · COMMUNITY FEED

Community benchmarks

What operators are actually running. Every row here was submitted by someone running runlocalai-bench, reviewed by editorial, and gated by reproduction status. Empty cells mean the submitter didn't report that number — never a fabrication.

SUBMIT YOUR RIG · TWO PATHS
Path A — directfastest, requires submit to API
One command, ships on completion
curl -fsSL https://runlocalai.co/bench.mjs |
  node --hardware <slug> --model <tag> --submit

5 runs × ~2 min. POSTs to /api/community/benchmarks on success.

Path B — review-firstpaste, review fields, submit
Run, then paste output here

Auto-parses 9 fields (model, cold-start, median, P5/P95, variance, OS, runtime, version, runs). Pick the matching hardware. Submit. 30 seconds total.

Open paste mode →

Both paths run the same protocol, hit the same review queue, and land here once approved. Anonymous OK. Source on /guides/methodology.

§ WHAT YOUR SUBMISSION POWERS

One measurement, eight surfaces upgraded

Every community benchmark that lands here replaces a bandwidth- derived estimate with a measured number across the site's flagships — same model × hardware × quant combo gets its confidence chip upgraded from E M.

GPU chooser

Measured tok/s on each hardware card replaces the extrapolated number

Quant advisor

The placeholder row turns into a reproduced measured row

Cost vs Cloud

Real tok/s drives the per-million-tokens local cost

Stream visualizer

Stream rate switches from estimate to measured

Compounder

Break-even months stop being approximations

Model Battle

Speed row on this hardware switches to measured

Hardware leaderboard

Score confidence chip upgrades from E → M for that row

Stack builder

Top-pick model recommendation switches to the actually-fastest

0 public submissions · newest first

WhenHardwareModelTok/sOperatorStatus
The feed is waiting for its first row.

Editorial benchmarks exist on the individual hardware pages, but community submissions through this feed haven't landed yet. Your run could be number one.

Submit a benchmark →
REVIEWED QUEUE

Editorial can approve a submission as plausible, but it stays out of this numeric feed until reproduction evidence lands.

REPRODUCED

We re-ran the same protocol on equivalent hardware and got within tolerance. Trustworthy.

INDEPENDENTLY REPRODUCED

Two or more independent operators filed matching numbers, and editorial confirmed. Highest confidence tier.