No machine on this network has measured this model yet. Everything below is what the catalog knows exists — where the weights come from, how large each build is, and whose work it is. Nothing here is a claim about what the model can do; that only appears once a machine has run it and signed the result.
unsloth/Kimi-K2-Thinking-GGUF:IQ1_Sllamacpp · IQ1_S · 265.7 GiB · quantizer (unsloth)
unsloth/Kimi-K2-Thinking-GGUF:IQ2_XXSllamacpp · IQ2_XXS · 312.2 GiB · quantizer (unsloth)
unsloth/Kimi-K2-Thinking-GGUF:IQ3_XXSllamacpp · IQ3_XXS · 392.7 GiB · quantizer (unsloth)
unsloth/Kimi-K2-Thinking-GGUF:Q4_0llamacpp · Q4_0 · 541.3 GiB · quantizer (unsloth)
moonshotai/Kimi-K2-Thinkingvllm · W4A16 · 553.4 GiB · vendor (moonshotai)
nvidia/Kimi-K2-Thinking-NVFP4vllm · NVFP4 · 553.4 GiB · quantizer (nvidia)
unsloth/Kimi-K2-Thinking-GGUF:Q4_K_Mllamacpp · Q4_K_M · 578.6 GiB · quantizer (unsloth)
mlx-community/Kimi-K2-Thinkingmlx · 4BIT · 612.5 GiB · quantizer (mlx-community)
unsloth/Kimi-K2-Thinking-GGUF:Q8_0llamacpp · Q8_0 · 1016.1 GiB · quantizer (unsloth)
unsloth/Kimi-K2-Thinking-GGUF:BF16llamacpp · BF16 · 1912.1 GiB · quantizer (unsloth)