family member

Llama 3.2 3B

13 builds of Llama 3.2 3B 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. 3.2B parameters, as the vendor states.

on ollama · 13 builds

trained 128K · proved 128K context proved by bartowski Q4_K_M

225.2 tok/s peak, by unsloth IQ1_S

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

buildpublisherprecisionenginewhose workcontext provedbest tok/smachines
hf.co/bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_MbartowskiQ4_K_Mollamaquantizer1,31,07286.91
hf.co/bartowski/Llama-3.2-3B-Instruct-GGUF:Q3_K_LbartowskiQ3_K_Lollamaquantizer1,31,07278.81
hf.co/bartowski/Llama-3.2-3B-Instruct-GGUF:IQ3_MbartowskiIQ3_Mollamaquantizer1,31,07272.01
hf.co/bartowski/Llama-3.2-3B-Instruct-GGUF:Q8_0bartowskiQ8_0ollamaquantizer1,31,07262.81
hf.co/bartowski/Llama-3.2-3B-Instruct-GGUF:F16bartowskiF16ollamaquantizer1,31,07238.31
hf.co/unsloth/Llama-3.2-3B-Instruct-GGUF:F16unslothF16ollamaquantizer16,38494.91
hf.co/unsloth/Llama-3.2-3B-Instruct-GGUF:IQ1_SunslothIQ1_Sollamaquantizer—225.22
hf.co/unsloth/Llama-3.2-3B-Instruct-GGUF:Q4_0unslothQ4_0ollamaquantizer—205.11
hf.co/unsloth/Llama-3.2-3B-Instruct-GGUF:Q4_K_MunslothQ4_K_Mollamaquantizer—203.91
hf.co/unsloth/Llama-3.2-3B-Instruct-GGUF:IQ3_XXSunslothIQ3_XXSollamaquantizer—199.51
hf.co/unsloth/Llama-3.2-3B-Instruct-GGUF:IQ2_XXSunslothIQ2_XXSollamaquantizer—124.21
hf.co/unsloth/Llama-3.2-3B-Instruct-GGUF:Q8_0unslothQ8_0ollamaquantizer—72.81
hf.co/MaziyarPanahi/Llama-3.2-3B-Instruct-GGUF:IQ1_SmaziyarpanahiIQ1_Sollama——45.71
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-3.2-3B-Instruct-uncensored1 build measuredLlama-3.2-3B-Instruct-Q4_K_M-GGUFnot measuredLlama-3.2-3B-Instructnot measuredLlama-3.2-3B-Instruct-abliterated-GGUFnot measuredLlama-3.2-3B-Instructnot measured