TensorFold community leaderboard
Benchmarks
The models, the machines, and the people setting the pace.
On the podium
Model capacity × measured speed from each contributor.
Loading the community podium
Across the community.
Self-reported standard runs. Models and hardware differ.
Across the community
Models on the board
Model size and speed
Publisher parameter estimates and measured token delivery.
Select a measured run to inspect its receipt.
Contributor profile
Loading contributor
Achievement cabinet
Earned from published benchmarks.
Display names are self-chosen; each upload identity has its own profile.
Explore the benchmarks
Every published run, newest first.
Latest runs
Select a model to inspect its receipt.
| Model | Hardware | Tokens / s | First text ms | Contributor | Receipt |
|---|
Run a benchmark
Chat, 256 output tokens, 5 repeats.
Upload community-v1 or current community-v2 receipts. Choose the matching protocol and topology in the filters to compare runs; their rankings stay separate.
tensorfold benchmark MODEL --output benchmark.jsonRun locally and keep the receipt. Uploads need no sign-in and accept community-v1 and current community-v2 receipts.