family member

Llama 3.2 1B

9 builds of Llama 3.2 1B measured on real machines — different packagers, precisions and engines, each a different file with its own numbers. The best each engine proved is first; every build has its own page with the machines and signatures behind it. 1.2B parameters, as the vendor states.

on ollama · 9 builds

proved 32K context proved by ollama Q8_0

113.3 tok/s peak, by MaziyarPanahi Q4_K_M

proved by at least one build: Tool calling · Tool loop finished · Faithful to tool results · Structured output · Sustained output · Writes code · Synthesises sources · Follows instructions

buildpublisherprecisionenginewhose workcontext provedbest tok/smachines
llama3.2:1bollamaQ8_0ollama—32,76849.910
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_MbartowskiQ4_K_Mollamaquantizer16,384103.94
hf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:IQ3_XSMaziyarPanahiIQ3_XSollamaquantizer4,096107.53
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q8_0bartowskiQ8_0ollamaquantizer4,0968.71
hf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:Q4_K_MMaziyarPanahiQ4_K_Mollamaquantizer—113.32
hf.co/MaziyarPanahi/Llama-3.2-1B-Instruct-GGUF:IQ1_SMaziyarPanahiIQ1_Sollamaquantizer—105.04
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q3_K_LbartowskiQ3_K_Lollamaquantizer—10.41
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:F16bartowskiF16ollamaquantizer—8.02
hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:IQ3_MbartowskiIQ3_Mollamaquantizer—5.31
derivatives — somebody else's change to these weights

Fine-tunes, merges and abliterations of this member, each its own model with its own page. None of their numbers are counted above.

Llama-3.2-1B-Instruct-Q4_K_M-GGUF1 build measuredLlama-3.2-1B-Instruct-bf16not measuredtofu_Llama-3.2-1B-Instruct_fullnot measuredLlama-3.2-1B-Instruct-LinkLlama-Cap50not measuredLlama-3.2-1B-Instructnot measured