Contributors › c4software's local-LLM benchmark configs
c4software's local-LLM benchmark configs
Every tracked benchmark config measured by c4software: model, quant, backend, tokens per second, source-linked and trust-tiered.
Snapshot 2026-10-15 · 394 configs on AI boxes and 423 on GPU cards · 123 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.
2 configs · 2 models · 1 stars · 75.9 tok/s best single-stream · c4software on github.
Configs measured by c4software
| Model | Hardware | Quant | Backend | Decode | Context | Trust | Source |
|---|---|---|---|---|---|---|---|
| Ornith-1.5-35B-A3B | Strix Halo | Q4_K_M | llama.cpp/rocm (speculative decode, mtp-ngram) | 75.9 tok/s (prefill 1306) | Extracted, source-linked | source | |
| Ornith-1.5-9B | Strix Halo | Q8_0 | llama.cpp/rocm (speculative decode, mtp-ngram) | 41.7 tok/s (prefill 1134) | Extracted, source-linked | source |
More: all configs · hardware · news · contributors · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.