Contributors › peonist-ai's local-LLM benchmark configs

peonist-ai's local-LLM benchmark configs

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

Snapshot 2026-09-30 · 254 configs · 116 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 · 763 stars · 46 tok/s best single-stream · peonist-ai on github.

Configs measured by peonist-ai

ModelQuantBackendDecodeContextTrustSource
Qwen3.8-Flash-NextW4Bhalogen/rocm (speculative decode, mtp)46 tok/s32768Extracted, source-linkedsource
Qwen3.8-Flash-NextW4Bhalogen/rocm (baseline)37.6 tok/s1500Extracted, source-linkedsource

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