Contributors › pepuscz's local-LLM benchmark configs

pepuscz's local-LLM benchmark configs

Every tracked benchmark config measured by pepuscz: model, quant, backend, tokens per second, source-linked and trust-tiered.

Snapshot 2026-09-10 · 148 configs · 69 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.

2 configs · 1 models · 13 stars · 35.7 tok/s best single-stream · pepuscz on github.

Configs measured by pepuscz

ModelQuantBackendDecodeContextTrustSource
DeepSeek-V4-Flash-0731UD-IQ3_XXSllama.cpp/vulkan35.7 tok/s (prefill 226.8)524288Extracted, source-linkedsource
DeepSeek-V4-Flash-0731UD-IQ3_XXSllama.cpp/rocm29.6 tok/s (prefill 142.8)131072Extracted, source-linkedsource

More: all configs · hardware · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.