Llama 3.1 8B Instruct › NVIDIA RTX 2070

Llama 3.1 8B Instruct on the NVIDIA RTX 2070

Best measured: 52.4 tokens/s generation and 1,252 tokens/s prompt processing (Q4_K_M, llamafile).

Every run

QuantBackendGenerate tok/sPrompt tok/sTestBuild / by / notesSource
Q4_K_Mllamafile 52.41,252LocalScore avg LocalScore rank 73/422LocalScore (localscore.ai)

tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128.

Check prices

Amazon Newegg

As an Amazon Associate I earn from qualifying purchases.

No card yet?

Try it on a rented GPU (RunPod)

Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.

All results for the NVIDIA RTX 2070 · Llama 3.1 8B Instruct on other hardware