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-K2.6-GGUF:Q2_K_XLllamacpp · Q2_K_XL · 317.1 GiB · quantizer (unsloth)
bartowski/moonshotai_Kimi-K2.6-GGUF:IQ3_XXSllamacpp · IQ3_XXS · 397.3 GiB · quantizer (bartowski)
amd/Kimi-K2.6-MXFP4vllm · QUARK · 520.6 GiB · quantizer (amd)
unsloth/Kimi-K2.6-GGUF:Q4_K_XLllamacpp · Q4_K_XL · 543.6 GiB · quantizer (unsloth)
unsloth/Kimi-K2.6-GGUF:Q8_K_XLllamacpp · Q8_K_XL · 553.7 GiB · quantizer (unsloth)
moonshotai/Kimi-K2.6vllm · BF16 · 554.3 GiB · vendor (moonshotai)
nvidia/Kimi-K2.6-NVFP4vllm · NVFP4 · 554.3 GiB · quantizer (nvidia)
DevQuasar/moonshotai.Kimi-K2.6-GGUF:Q4_K_Mllamacpp · Q4_K_M · 577.9 GiB · quantizer (DevQuasar)
aidiffuser/Kimi-K2.6-MLX-4bitmlx · 4BIT · 612.5 GiB · quantizer (aidiffuser)
RedHatAI/Kimi-K2.6-FP8-BLOCKvllm · FP8 · 960.1 GiB · quantizer (RedHatAI)
DevQuasar/moonshotai.Kimi-K2.6-GGUF:Q8_0llamacpp · Q8_0 · 1016.1 GiB · quantizer (DevQuasar)
unsloth/Kimi-K2.6-GGUF:BF16llamacpp · BF16 · 1912.1 GiB · quantizer (unsloth)