node_a414d806 · 2026-09-27 21:56 UTC · Chrome on Windows
First GTX 1060 on the network
The reference machine scores 1,000. This one delivers 1.13× its throughput.
This is an entry-level machine.
EntryFine for small models. Larger ones will feel slow.
71 points from Mid-range.
The network has 3 verified runs. This comparison opens at 10.
1 verified run in this class so far. The comparison opens once three others exist.
Loses a little speed as it heats up.
Streaming loads. The path every token takes through the weights.
Streaming stores. KV-cache appends and activations.
Buffer to buffer. Read and write traffic together.
Scattered gathers. Embedding lookups and sparse access.
Dense matrix multiply. The arithmetic inside every layer.
Continuous load. What the machine holds once it heats up.
- Llama 3.2 3B3.2 GB≤ 77.1 tok/s
- Llama 3.1 8B6.4 GBdoes not fit
- Llama 3.3 70B45.9 GBdoes not fit
- Phi-410.7 GBdoes not fit
- Gemma 2 27B18.7 GBdoes not fit
- Qwen 2.5 32B22.2 GBdoes not fit
4-bit weights, 8K context. The ceiling is read throughput divided by the weights read per token; real runtimes land below it.
- Verified run+100
- First run on GTX 1060+100
- Early data for this class+250
- Genesis wafer+500
- The tab was hidden for part of the run.
Deep report
$XCOMP holdersEvery phase sample by sample, the spread between the 5th and 95th percentile, this run against its class median, and a full export.
How does your machine compare?
Forty-five seconds, in this browser.