GB10

106 records across 13 models (quants: BF16, IQ2_XXS, IQ2_XXS+Q5_K, NVFP4, Q4_0, Q4_K - Medium, Q4_K_M, Q8_0, UD-IQ4_XS, UD-Q4_K_M, UD-Q8_K_L).

hardwaremodelquantbackendtok/spp tok/sWttft sctx/tgdateclassflagssource
GB10Falcon-H1-7B 7BQ4_K_Mllama.cpp (CUDA)273.1––––2026-08community–source
GB10Falcon-H1-7B 7BBF16llama.cpp (CUDA)199.0––––2026-08community–source
GB10Falcon-H1-7B 7BQ4_K_Mllama.cpp (CUDA)187.7––––2026-08community–source
GB10Falcon-H1-7B 7BBF16llama.cpp (CUDA)80.0––––2026-08community–source
GB10Gemma 4 12B Q4_K_Mllama.cpp68.42––––2026-10-12sourced–source
GB10Gemma 4 12B Q4_K_Mllama.cpp67.31––––2026-10-12sourced–source
GB10GLM-4.6 IQ2_XXS+Q5_Kllama.cpp (CUDA)17.14–––1136642026-10-03community–source
GB10GLM-4.6 IQ2_XXS+Q5_Kllama.cpp (CUDA)14.36–––327682026-10-03community–source
GB10GLM-4.6 IQ2_XXS+Q5_Kllama.cpp (CUDA)12.17–––327682026-10-03community–source
GB10GLM-4.6 IQ2_XXS+Q5_Kllama.cpp (CUDA)11.45–––327682026-10-03community–source
GB10GLM-4.6 IQ2_XXS+Q5_Kllama.cpp (CUDA)4.10197––576002026-10-03community–source
GB10GLM-4.6 IQ2_XXS+Q5_Kllama.cpp (CUDA)3.61125––1119002026-10-03community–source
GB10granite-4.0-h-tiny Q4_K_Mllama.cpp (CUDA)551.2––––2026-08community–source
GB10granite-4.0-h-tiny Q4_K_Mllama.cpp (CUDA)456.4––––2026-08community–source
GB10granite-4.0-h-tiny BF16llama.cpp (CUDA)390.4––––2026-08community–source
GB10granite-4.0-h-tiny BF16llama.cpp (CUDA)271.3––––2026-08community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)60.393321.91––tg1282026-07-14community–source
GB10llama 7B Q4_0llama.cpp (CUDA)60.243310.48––tg1282026-01-20community–source
GB10llama 7B Q4_0llama.cpp (CUDA)59.624016.46––tg1282026-03-30community–source
GB10llama 7B Q4_0llama.cpp (CUDA)59.063306.61––tg1282026-03-30community–source
GB10llama 7B Q4_0llama.cpp (CUDA)59.053954.97––tg1282026-03-30community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)58.512789.51––tg1282026-07-14community–source
GB10llama 7B Q4_0llama.cpp (CUDA)58.383261.99––tg1282026-03-30community–source
GB10llama 7B Q4_0llama.cpp (CUDA)58.143861.19––tg1282026-02-16community–source
GB10llama 7B Q4_0llama.cpp (CUDA)58.133241.51––tg1282026-02-16community–source
GB10llama 7B Q4_0llama.cpp (CUDA)57.213062.31––tg1282025-10-15community–source
GB10llama 7B Q4_0llama.cpp (CUDA)56.743661.37––tg1282025-10-15community–source
GB10llama 7B Q4_0llama.cpp (CUDA)54.913920.63––tg1282026-03-30community–source
GB10llama 7B Q4_0llama.cpp (CUDA)54.393293.78––tg1282026-03-30community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)53.643279.89––tg1282026-03-15community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)53.213219.57––tg1282026-03-15community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)52.282737.79––tg1282026-03-15community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)51.72710.58––tg1282026-03-15community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)50.82665.61––tg1282026-01-20community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)48.422692.06––tg1282026-03-15community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)47.552731.93––tg1282026-03-14community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)46.673137.97––tg1282026-03-15community–source
GB10llama 7B Q4_0llama.cpp (Vulkan)46.613171.16––tg1282026-03-14community–source
GB10llama 7B Q4_0llama.cpp (CPU)24.08313.01––tg1282026-01-20community–source
GB10mamba2-2.7b 2.7BQ8_0llama.cpp (CUDA)246.7––––2026-08community–source
GB10mamba2-2.7b 2.7BQ8_0llama.cpp (CUDA)98.4––––2026-08community–source
GB10Nemotron 3 Nano 30B-A3B 30BNVFP4llama.cpp (CUDA)660.8––––2026-08community–source
GB10Nemotron 3 Nano 30B-A3B 30BNVFP4llama.cpp (CUDA)350.8––––2026-08community–source
GB10Nemotron 3 Nano 30B-A3B 30BNVFP4llama.cpp (CUDA)292.5––––2026-08community–source
GB10Nemotron 3 Nano 30B-A3B 30BNVFP4llama.cpp (CUDA)214.8––––2026-08community–source
GB10Nemotron 3 Nano 30B-A3B 30BNVFP4llama.cpp (CUDA)202.6––––2026-08community–source
GB10Nemotron 3 Nano 30B-A3B 30BNVFP4llama.cpp (CUDA)152.4––––2026-08community–source
GB10Nemotron 3 Nano 30B-A3B 30BNVFP4llama.cpp (CUDA)67.7––––2026-08community–source
GB10Nemotron 3 Nano 30B-A3B 30BNVFP4llama.cpp (CUDA)67.5––––2026-08community–source
GB10nemotron_h_moe 31B.A3.5B Q4_K - Mediumllama.cpp (Vulkan)64.51661.48––tg1282026-01-17community–source
GB10nemotron_h_moe 31B.A3.5B Q4_K - Mediumllama.cpp (CUDA)61.762135.98––tg1282026-01-17community–source
GB10Ornith-1.5-397B-A17B IQ2_XXSllama.cpp (CUDA)19.82–––327682026-10-02community–source
GB10Ornith-1.5-397B-A17B IQ2_XXSllama.cpp (CUDA)19.18–––327682026-10-02community–source
GB10Ornith-1.5-397B-A17B IQ2_XXSllama.cpp (CUDA)19.07–––327682026-10-02community–source
GB10Ornith-1.5-397B-A17B IQ2_XXSllama.cpp (CUDA)17.72–––327682026-10-02community–source
GB10Ornith-1.5-397B-A17B IQ2_XXSllama.cpp (CUDA)15.00–––327682026-10-02community–source
GB10Qwen3.6-27B Q4_K_Mllama.cpp26.77––––2026-10-12sourced–source
GB10Qwen3.6-27B Q4_K_Mllama.cpp24.29––––2026-10-12sourced–source
GB10Qwen3.6-35B-A3B NVFP4vLLM118.1–––tg2562026-10-06community–source
GB10Qwen3.6-35B-A3B NVFP4vLLM72.9–––tg2562026-10-06community–source
GB10Qwen3.6-35B-A3B UD-Q4_K_Mllama.cpp–3271.18––20482026-10-13sourced–source
GB10Qwen3.6-35B-A3B UD-Q4_K_Mllama.cpp–3406.68––20482026-10-13sourced–source
GB10Qwen3.6-35B-A3B UD-Q4_K_Mllama.cpp–3254.79––40962026-10-13sourced–source
GB10Qwen3.6-35B-A3B UD-Q4_K_Mllama.cpp–3366.88––40962026-10-13sourced–source
GB10Qwen3.8-27B UD-Q8_K_Lllama.cpp–942.84––20482026-10-13sourced–source
GB10Qwen3.8-27B UD-Q8_K_Lllama.cpp–984.86––20482026-10-13sourced–source
GB10Qwen3.8-27B UD-Q8_K_Lllama.cpp–928.16––40962026-10-13sourced–source
GB10Qwen3.8-27B UD-Q8_K_Lllama.cpp–973.34––40962026-10-13sourced–source
GB10Qwen3.8-Flash-Next NVFP4vLLM73.43–––tg2562026-10-02community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM72.64–––tg2562026-10-02community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM67.4––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM66.2––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM60.9––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM58.5––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM55.28–––tg2562026-10-02community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM54.28–––tg2562026-10-02community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM52.5––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM51.8––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM51.5––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM51.1––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM50.5–––tg2562026-10-12community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM49.1–––tg2562026-10-12community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM48.14–––tg2562026-10-02community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM47.70–––tg2562026-10-02community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM46.5––––2026-10-04community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM46.32–––tg2562026-10-02community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM45.92––7.12409602026-09-10sourced–source
GB10Qwen3.8-Flash-Next NVFP4vLLM44.86–––tg2562026-10-02community–source
GB10Qwen3.8-Flash-Next UD-IQ4_XSllama.cpp43.97––––2026-10-12sourced–source
GB10Qwen3.8-Flash-Next NVFP4vLLM43.44––6.75409602026-09-10sourced–source
GB10Qwen3.8-Flash-Next NVFP4vLLM42.84––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM42.07––––2026-10-08community–source
GB10Qwen3.8-Flash-Next UD-IQ4_XSllama.cpp41.51––––2026-10-12sourced–source
GB10Qwen3.8-Flash-Next NVFP4vLLM41.06––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM40.76––––2026-10-08community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM39.4––––2026-10-04community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM36.31––5.44409602026-09-10sourced–source
GB10Qwen3.8-Flash-Next NVFP4vLLM34.25––5.09409602026-09-10sourced–source
GB10Qwen3.8-Flash-Next NVFP4vLLM33.0–––tg2562026-10-12community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM32.6–––tg2562026-10-12community–source
GB10Qwen3.8-Flash-Next NVFP4vLLM32.41––3.4409602026-09-10sourced–source
GB10Qwen3.8-Flash-Next NVFP4vLLM23.51––3.45409602026-09-10sourced–source
GB10Qwen3.8-Flash-Next UD-IQ4_XSllama.cpp–1044.71––20482026-10-13sourced–source
GB10Qwen3.8-Flash-Next UD-IQ4_XSllama.cpp–1104.31––20482026-10-13sourced–source
GB10Qwen3.8-Flash-Next UD-IQ4_XSllama.cpp–1068.7––40962026-10-13sourced–source
GB10Qwen3.8-Flash-Next UD-IQ4_XSllama.cpp–1124.01––40962026-10-13sourced–source

All rows cite their source. Provenance: community 40018 sourced 1085

← all hardware