llamaperf

NInfer benchmarks

An inference engine for running open-weight LLMs locally.

47 community reports

NInfer results by GPU

Hardware with reported NInfer results, with each report's model, quant and speed, newest first. Runs on several cards, with speculative decoding, with batched requests or with part of the model in system RAM say so, since each describes a different setup. Prompt processing is how fast the input is read, not how long you wait for the first word. How to measure time to first token.

NInfer on NVIDIA RTX 509018 reports

All NInfer results on the NVIDIA RTX 5090Filter NInfer reports on the NVIDIA RTX 5090

NInfer on NVIDIA RTX 40906 reports

All NInfer results on the NVIDIA RTX 4090Filter NInfer reports on the NVIDIA RTX 4090

NInfer on NVIDIA V100 16GB3 reports

Filter NInfer reports on the NVIDIA V100 16GB

NInfer on NVIDIA RTX 30902 reports

Filter NInfer reports on the NVIDIA RTX 3090

NInfer on NVIDIA RTX 5060 Ti 16GB2 reports

Filter NInfer reports on the NVIDIA RTX 5060 Ti 16GB

NInfer, GPU not identified1 report

This engine doesn't yet have an editorial profile on llamaperf. The community reports below show how it's been used in practice across different hardware.

Compare GPUs running NInfer

Every card is compared on one model, Qwen3.8 125B · 6B active at 4-bit, the size and bit class that most cards here have plain runs of. A card with no run of that model shows none rather than a figure from a different model.

GPUVRAMReportsMedian speed, Qwen3.8 125B · 6B active 4-bit
NVIDIA RTX 5090nvidia32GB23no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX 4090nvidia24GB7no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX Pro 6000 Blackwellnvidia96GB4Gen 172.0 t/sPP 5905.0 t/s · n=1
NVIDIA RTX 5060 Ti 16GBnvidia16GB3no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX 5080nvidia16GB3no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA V100 16GBnvidia16GB3no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX 2080 Ti 22GB (modded)nvidia22GB2no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX 3090nvidia24GB2no Qwen3.8 125B · 6B active 4-bit run reported

NInfer against other engines

Pairs of plain runs on the same card, of the same model size at the same bit class: one device, one request, no speculative decoding, the whole model in memory. Context length and build still differ between the two sides, and each side shows its own.

Model and GPUNInferOther engine
Qwen3.8 125B · 6B active 4-bitNVIDIA RTX Pro 6000 Blackwell172.0 t/sPrompt processing (prefill) 5905.0 tokens/s (1)one generation report, 512 contextik_llama.cpp40.0 t/sPrompt processing (prefill) 2000.0 tokens/s (1)one generation report, 200K context
Qwen3.8 125B · 6B active 4-bitNVIDIA RTX Pro 6000 Blackwell172.0 t/sPrompt processing (prefill) 5905.0 tokens/s (1)one generation report, 512 contextllama.cpp105.5 t/sPrompt processing (prefill) 1955.0 tokens/s (1)median of 2 generation reports, 2K to 240K context
Qwen3.8 125B · 6B active 4-bitNVIDIA RTX Pro 6000 Blackwell172.0 t/sPrompt processing (prefill) 5905.0 tokens/s (1)one generation report, 512 contextSGLang-Prompt processing (prefill) 11352.0 tokens/s (1)no generation reports, 256K context

Top models on NInfer

Frequently asked

Is NInfer faster than other engines?

llamaperf has 2 matched pairs of NInfer and other-engine runs on the same card, model size and bit class, but none is comparable enough to say which engine is faster: the sides have too few reports or were run at different context lengths. A pair counts toward a direction only when each side has at least 3 generation reports with a stated context, and the two sides' contexts overlap. The measured figures and their conditions are in the table on this page.