llamaperf

ik_llama.cpp benchmarks

An inference engine for running open-weight LLMs locally.

12 community reports

ik_llama.cpp results by GPU

Hardware with reported ik_llama.cpp 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.

ik_llama.cpp, GPU not identified8 reports

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 ik_llama.cpp

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 3090nvidia24GB3no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX 5070 Tinvidia16GB2no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX Pro 6000 Blackwellnvidia96GB2Gen 40.0 t/sPP 2000.0 t/s · n=1
NVIDIA RTX 3060 12GBnvidia12GB1no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX 4070nvidia12GB1no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX 4070 Supernvidia12GB1no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA RTX 5090nvidia32GB1no Qwen3.8 125B · 6B active 4-bit run reported
NVIDIA T4 16GBnvidia16GB1no Qwen3.8 125B · 6B active 4-bit run reported

ik_llama.cpp 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 GPUik_llama.cppOther engine
Qwen3.8 125B · 6B active 4-bitNVIDIA RTX Pro 6000 Blackwell40.0 t/sPrompt processing (prefill) 2000.0 tokens/s (1)one generation report, 200K 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 Blackwell40.0 t/sPrompt processing (prefill) 2000.0 tokens/s (1)one generation report, 200K contextNInfer172.0 t/sPrompt processing (prefill) 5905.0 tokens/s (1)one generation report, 512 context
Qwen3.8 125B · 6B active 4-bitNVIDIA RTX Pro 6000 Blackwell40.0 t/sPrompt processing (prefill) 2000.0 tokens/s (1)one generation report, 200K contextSGLang-Prompt processing (prefill) 11352.0 tokens/s (1)no generation reports, 256K context

Top models on ik_llama.cpp

Frequently asked

Is ik_llama.cpp faster than other engines?

llamaperf has 2 matched pairs of ik_llama.cpp 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.