{"meta":{"apiVersion":"v1","generatedAt":"2026-10-19T06:51:43.218Z","source":"live","collectedAt":null,"docs":"https://aicomputeradar.dev/developers","openapi":"https://aicomputeradar.dev/api/v1/openapi.json","license":{"license":"CC BY 4.0","licenseUrl":"https://creativecommons.org/licenses/by/4.0/","attribution":"AI Compute Radar (https://aicomputeradar.dev)","attributionRequired":true,"upstream":["Hugging Face Hub public model metadata (trending, downloads, likes)","OpenRouter model catalogue (CC BY 4.0)","Vast.ai public marketplace offers (aggregated)","RunPod posted on-demand prices","Clore.ai public marketplace offers (aggregated)","AWS public on-demand price list","Azure retail prices API","Oracle Cloud price list","Vendor product documentation (hardware facts)"]}},"data":{"hardware":{"id":"rtx-4090","name":"RTX 4090","memoryGb":24,"memoryArchitecture":"discrete","category":"high_vram_consumer_gpu","usableMemoryGb":21.1,"usableMemoryNote":null,"memoryBandwidthGbps":1008,"page":"https://aicomputeradar.dev/local/gpu/rtx-4090"},"assumptions":{"contextTokens":16384,"kvCacheType":"f16"},"best":{"model":{"slug":"gemma-4-26b-a4b","name":"Gemma 4 26B-A4B","creator":"Google"},"quantization":"Q4_K_M","label":"GOOD","score":74,"confidence":"high","estimateGb":{"weights":15.8,"contextCache":0.82,"overhead":1.6,"total":18.2,"usableMemory":21.1,"offload":0},"utilization":0.861,"explanation":["Weights ~15.8 GB (Q4_K_M), context ~0.8 GB at 16K tokens, runtime ~1.6 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Fits with room to spare at this context length."],"links":{"gguf":"https://huggingface.co/unsloth/gemma-4-26B-A4B-it-GGUF","huggingface":"https://huggingface.co/google/gemma-4-26B-A4B-it"}},"fits":[{"model":{"slug":"gpt-oss-20b","name":"gpt-oss 20B","creator":"OpenAI"},"quantization":"Q4_K_M","label":"EXCELLENT","score":82,"confidence":"high","estimateGb":{"weights":10.8,"contextCache":0.38,"overhead":1.4,"total":12.6,"usableMemory":21.1,"offload":0},"utilization":0.597,"explanation":["Weights ~10.8 GB (Q4_K_M), context ~0.4 GB at 16K tokens, runtime ~1.4 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/unsloth/gpt-oss-20b-GGUF","huggingface":"https://huggingface.co/openai/gpt-oss-20b"}},{"model":{"slug":"qwen3-14b","name":"Qwen3 14B","creator":"Alibaba Qwen"},"quantization":"Q4_K_M","label":"EXCELLENT","score":83,"confidence":"high","estimateGb":{"weights":8.4,"contextCache":2.5,"overhead":1.4,"total":12.3,"usableMemory":21.1,"offload":0},"utilization":0.58,"explanation":["Weights ~8.4 GB (Q4_K_M), context ~2.5 GB at 16K tokens, runtime ~1.4 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/Qwen/Qwen3-14B-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3-14B"}},{"model":{"slug":"gemma-4-12b","name":"Gemma 4 12B","creator":"Google"},"quantization":"Q4_K_M","label":"EXCELLENT","score":87,"confidence":"high","estimateGb":{"weights":6.6,"contextCache":1.31,"overhead":1.3,"total":9.2,"usableMemory":21.1,"offload":0},"utilization":0.436,"explanation":["Weights ~6.6 GB (Q4_K_M), context ~1.3 GB at 16K tokens, runtime ~1.3 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/unsloth/gemma-4-12B-it-GGUF","huggingface":"https://huggingface.co/google/gemma-4-12B-it"}},{"model":{"slug":"nemotron-nano-9b","name":"Nemotron Nano 9B v2","creator":"NVIDIA"},"quantization":"Q4_K_M","label":"EXCELLENT","score":85,"confidence":"high","estimateGb":{"weights":6.1,"contextCache":3.5,"overhead":1.3,"total":10.9,"usableMemory":21.1,"offload":0},"utilization":0.515,"explanation":["Weights ~6.1 GB (Q4_K_M), context ~3.5 GB at 16K tokens, runtime ~1.3 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/bartowski/nvidia_NVIDIA-Nemotron-Nano-9B-v2-GGUF","huggingface":"https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-9B-v2"}},{"model":{"slug":"ornith-1-5-9b","name":"Ornith 1.5 9B","creator":"Ornith AI"},"quantization":"Q4_K_M","label":"EXCELLENT","score":90,"confidence":"high","estimateGb":{"weights":5.4,"contextCache":0.5,"overhead":1.3,"total":7.2,"usableMemory":21.1,"offload":0},"utilization":0.339,"explanation":["Weights ~5.4 GB (Q4_K_M), context ~0.5 GB at 16K tokens, runtime ~1.3 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/ornith-ai/Ornith-1.5-9B-GGUF","huggingface":"https://huggingface.co/ornith-ai/Ornith-1.5-9B"}},{"model":{"slug":"granite-4-2-8b","name":"Granite 4.2 8B","creator":"IBM"},"quantization":"Q4_K_M","label":"EXCELLENT","score":88,"confidence":"high","estimateGb":{"weights":5,"contextCache":2.5,"overhead":1.3,"total":8.8,"usableMemory":21.1,"offload":0},"utilization":0.414,"explanation":["Weights ~5.0 GB (Q4_K_M), context ~2.5 GB at 16K tokens, runtime ~1.3 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/ibm-granite/granite-4.2-8b-GGUF","huggingface":"https://huggingface.co/ibm-granite/granite-4.2-8b"}},{"model":{"slug":"lfm2-5-8b-a1b","name":"LFM2.5 8B-A1B","creator":"Liquid AI"},"quantization":"Q4_K_M","label":"EXCELLENT","score":91,"confidence":"high","estimateGb":{"weights":4.8,"contextCache":0.19,"overhead":1.2,"total":6.2,"usableMemory":21.1,"offload":0},"utilization":0.295,"explanation":["Weights ~4.8 GB (Q4_K_M), context ~0.2 GB at 16K tokens, runtime ~1.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/LiquidAI/LFM2.5-8B-A1B-GGUF","huggingface":"https://huggingface.co/LiquidAI/LFM2.5-8B-A1B"}},{"model":{"slug":"qwen3-8b","name":"Qwen3 8B","creator":"Alibaba Qwen"},"quantization":"Q4_K_M","label":"EXCELLENT","score":88,"confidence":"high","estimateGb":{"weights":4.7,"contextCache":2.25,"overhead":1.2,"total":8.2,"usableMemory":21.1,"offload":0},"utilization":0.388,"explanation":["Weights ~4.7 GB (Q4_K_M), context ~2.3 GB at 16K tokens, runtime ~1.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/Qwen/Qwen3-8B-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3-8B"}},{"model":{"slug":"gemma-4-e4b","name":"Gemma 4 E4B","creator":"Google"},"quantization":"Q4_K_M","label":"EXCELLENT","score":91,"confidence":"high","estimateGb":{"weights":4.6,"contextCache":0.25,"overhead":1.2,"total":6.1,"usableMemory":21.1,"offload":0},"utilization":0.288,"explanation":["Weights ~4.6 GB (Q4_K_M), context ~0.3 GB at 16K tokens, runtime ~1.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/unsloth/gemma-4-E4B-it-GGUF","huggingface":"https://huggingface.co/google/gemma-4-E4B-it"}},{"model":{"slug":"ling-3-0-tiny","name":"Ling 3.0 Tiny","creator":"InclusionAI"},"quantization":"Q4_K_M","label":"EXCELLENT","score":91,"confidence":"high","estimateGb":{"weights":4.5,"contextCache":0.42,"overhead":1.2,"total":6.2,"usableMemory":21.1,"offload":0},"utilization":0.292,"explanation":["Weights ~4.5 GB (Q4_K_M), context ~0.4 GB at 16K tokens, runtime ~1.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/bloomer010/Ling-3.0-tiny-GGUF","huggingface":"https://huggingface.co/inclusionAI/Ling-3.0-tiny"}},{"model":{"slug":"qwen3-4b","name":"Qwen3 4B","creator":"Alibaba Qwen"},"quantization":"Q4_K_M","label":"EXCELLENT","score":92,"confidence":"high","estimateGb":{"weights":2.3,"contextCache":2.25,"overhead":1.2,"total":5.7,"usableMemory":21.1,"offload":0},"utilization":0.271,"explanation":["Weights ~2.3 GB (Q4_K_M), context ~2.3 GB at 16K tokens, runtime ~1.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/unsloth/Qwen3-4B-Instruct-2507-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507"}},{"model":{"slug":"phi-4-mini","name":"Phi-4 Mini","creator":"Microsoft"},"quantization":"Q4_K_M","label":"EXCELLENT","score":92,"confidence":"high","estimateGb":{"weights":2.3,"contextCache":2,"overhead":1.2,"total":5.5,"usableMemory":21.1,"offload":0},"utilization":0.259,"explanation":["Weights ~2.3 GB (Q4_K_M), context ~2.0 GB at 16K tokens, runtime ~1.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/unsloth/Phi-4-mini-instruct-GGUF","huggingface":"https://huggingface.co/microsoft/Phi-4-mini-instruct"}},{"model":{"slug":"granite-4-2-3b","name":"Granite 4.2 3B","creator":"IBM"},"quantization":"Q4_K_M","label":"EXCELLENT","score":94,"confidence":"high","estimateGb":{"weights":2.1,"contextCache":1.25,"overhead":1.2,"total":4.5,"usableMemory":21.1,"offload":0},"utilization":0.214,"explanation":["Weights ~2.1 GB (Q4_K_M), context ~1.3 GB at 16K tokens, runtime ~1.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/ibm-granite/granite-4.2-3b-GGUF","huggingface":"https://huggingface.co/ibm-granite/granite-4.2-3b"}},{"model":{"slug":"lfm2-5-2-6b","name":"LFM2.5 2.6B","creator":"Liquid AI"},"quantization":"Q4_K_M","label":"EXCELLENT","score":96,"confidence":"high","estimateGb":{"weights":1.6,"contextCache":0.25,"overhead":1.1,"total":3,"usableMemory":21.1,"offload":0},"utilization":0.142,"explanation":["Weights ~1.6 GB (Q4_K_M), context ~0.3 GB at 16K tokens, runtime ~1.1 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/LiquidAI/LFM2.5-2.6B-GGUF","huggingface":"https://huggingface.co/LiquidAI/LFM2.5-2.6B"}},{"model":{"slug":"lfm2-5-1-2b","name":"LFM2.5 1.2B","creator":"Liquid AI"},"quantization":"Q4_K_M","label":"EXCELLENT","score":97,"confidence":"high","estimateGb":{"weights":0.7,"contextCache":0.19,"overhead":1.1,"total":2,"usableMemory":21.1,"offload":0},"utilization":0.095,"explanation":["Weights ~0.7 GB (Q4_K_M), context ~0.2 GB at 16K tokens, runtime ~1.1 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct-GGUF","huggingface":"https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct"}},{"model":{"slug":"qwen3-0-6b","name":"Qwen3 0.6B","creator":"Alibaba Qwen"},"quantization":"Q8_0","label":"EXCELLENT","score":95,"confidence":"high","estimateGb":{"weights":0.6,"contextCache":1.75,"overhead":1.1,"total":3.5,"usableMemory":21.1,"offload":0},"utilization":0.164,"explanation":["Weights ~0.6 GB (Q8_0), context ~1.8 GB at 16K tokens, runtime ~1.1 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Comfortable headroom — you can raise the context length or run other apps alongside it."],"links":{"gguf":"https://huggingface.co/Qwen/Qwen3-0.6B-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3-0.6B"}},{"model":{"slug":"gemma-4-26b-a4b","name":"Gemma 4 26B-A4B","creator":"Google"},"quantization":"Q4_K_M","label":"GOOD","score":74,"confidence":"high","estimateGb":{"weights":15.8,"contextCache":0.82,"overhead":1.6,"total":18.2,"usableMemory":21.1,"offload":0},"utilization":0.861,"explanation":["Weights ~15.8 GB (Q4_K_M), context ~0.8 GB at 16K tokens, runtime ~1.6 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Fits with room to spare at this context length."],"links":{"gguf":"https://huggingface.co/unsloth/gemma-4-26B-A4B-it-GGUF","huggingface":"https://huggingface.co/google/gemma-4-26B-A4B-it"}},{"model":{"slug":"gemma-3-27b","name":"Gemma 3 27B","creator":"Google"},"quantization":"Q4_K_M","label":"GOOD","score":74,"confidence":"high","estimateGb":{"weights":15.4,"contextCache":1.66,"overhead":1.6,"total":18.6,"usableMemory":21.1,"offload":0},"utilization":0.882,"explanation":["Weights ~15.4 GB (Q4_K_M), context ~1.7 GB at 16K tokens, runtime ~1.6 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Fits with room to spare at this context length."],"links":{"gguf":"https://huggingface.co/unsloth/gemma-3-27b-it-GGUF","huggingface":"https://huggingface.co/google/gemma-3-27b-it"}},{"model":{"slug":"qwen3-8-27b","name":"Qwen3.8 27B","creator":"Alibaba Qwen"},"quantization":"Q4_K_M","label":"GOOD","score":75,"confidence":"high","estimateGb":{"weights":15.3,"contextCache":1,"overhead":1.6,"total":17.9,"usableMemory":21.1,"offload":0},"utilization":0.846,"explanation":["Weights ~15.3 GB (Q4_K_M), context ~1.0 GB at 16K tokens, runtime ~1.6 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Fits with room to spare at this context length."],"links":{"gguf":"https://huggingface.co/unsloth/Qwen3.8-27B-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3.8-27B"}},{"model":{"slug":"mistral-small-3-2","name":"Mistral Small 3.2","creator":"Mistral AI"},"quantization":"Q4_K_M","label":"GOOD","score":75,"confidence":"high","estimateGb":{"weights":13.3,"contextCache":2.5,"overhead":1.5,"total":17.3,"usableMemory":21.1,"offload":0},"utilization":0.819,"explanation":["Weights ~13.3 GB (Q4_K_M), context ~2.5 GB at 16K tokens, runtime ~1.5 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Fits with room to spare at this context length."],"links":{"gguf":"https://huggingface.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF","huggingface":"https://huggingface.co/mistralai/Mistral-Small-3.2-24B-Instruct-2506"}},{"model":{"slug":"devstral-small-2-24b","name":"Devstral Small 2 24B","creator":"Mistral AI"},"quantization":"Q4_K_M","label":"GOOD","score":75,"confidence":"high","estimateGb":{"weights":13.3,"contextCache":2.5,"overhead":1.5,"total":17.3,"usableMemory":21.1,"offload":0},"utilization":0.819,"explanation":["Weights ~13.3 GB (Q4_K_M), context ~2.5 GB at 16K tokens, runtime ~1.5 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Fits with room to spare at this context length."],"links":{"gguf":"https://huggingface.co/unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF","huggingface":"https://huggingface.co/mistralai/Devstral-Small-2-24B-Instruct-2512"}},{"model":{"slug":"qwen3-coder-30b","name":"Qwen3-Coder 30B-A3B","creator":"Alibaba Qwen"},"quantization":"Q4_K_M","label":"TIGHT","score":71,"confidence":"high","estimateGb":{"weights":17.3,"contextCache":1.5,"overhead":1.6,"total":20.4,"usableMemory":21.1,"offload":0},"utilization":0.967,"explanation":["Weights ~17.3 GB (Q4_K_M), context ~1.5 GB at 16K tokens, runtime ~1.6 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Fits, but with little headroom. Lower the context or quantize the KV cache if it stutters."],"links":{"gguf":"https://huggingface.co/unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct"}},{"model":{"slug":"glm-4-7-flash","name":"GLM-4.7-Flash","creator":"Zhipu AI"},"quantization":"Q4_K_M","label":"TIGHT","score":72,"confidence":"high","estimateGb":{"weights":17.1,"contextCache":0.83,"overhead":1.6,"total":19.5,"usableMemory":21.1,"offload":0},"utilization":0.925,"explanation":["Weights ~17.1 GB (Q4_K_M), context ~0.8 GB at 16K tokens, runtime ~1.6 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Fits, but with little headroom. Lower the context or quantize the KV cache if it stutters."],"links":{"gguf":"https://huggingface.co/unsloth/GLM-4.7-Flash-GGUF","huggingface":"https://huggingface.co/zai-org/GLM-4.7-Flash"}},{"model":{"slug":"ling-3-0-flash","name":"Ling 3.0 Flash","creator":"InclusionAI"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":0,"confidence":"high","estimateGb":{"weights":70.1,"contextCache":0.74,"overhead":3.2,"total":74,"usableMemory":21.1,"offload":52.9},"utilization":3.506,"explanation":["Weights ~70.1 GB (Q4_K_M), context ~0.7 GB at 16K tokens, runtime ~3.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 52.9 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/AtomicChat/Ling-3.0-flash-GGUF","huggingface":"https://huggingface.co/inclusionAI/Ling-3.0-flash"}},{"model":{"slug":"glm-4-5-air","name":"GLM-4.5-Air","creator":"Zhipu AI"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":0,"confidence":"high","estimateGb":{"weights":68,"contextCache":2.88,"overhead":3.1,"total":74,"usableMemory":21.1,"offload":52.9},"utilization":3.504,"explanation":["Weights ~68.0 GB (Q4_K_M), context ~2.9 GB at 16K tokens, runtime ~3.1 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 52.9 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/unsloth/GLM-4.5-Air-GGUF","huggingface":"https://huggingface.co/zai-org/GLM-4.5-Air"}},{"model":{"slug":"gpt-oss-120b","name":"gpt-oss 120B","creator":"OpenAI"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":0,"confidence":"high","estimateGb":{"weights":58.5,"contextCache":0.57,"overhead":2.9,"total":61.9,"usableMemory":21.1,"offload":40.8},"utilization":2.932,"explanation":["Weights ~58.5 GB (Q4_K_M), context ~0.6 GB at 16K tokens, runtime ~2.9 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 40.8 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/unsloth/gpt-oss-120b-GGUF","huggingface":"https://huggingface.co/openai/gpt-oss-120b"}},{"model":{"slug":"qwen3-coder-next","name":"Qwen3-Coder Next","creator":"Alibaba Qwen"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":0,"confidence":"high","estimateGb":{"weights":45.1,"contextCache":1.5,"overhead":2.5,"total":49.1,"usableMemory":21.1,"offload":27.9},"utilization":2.323,"explanation":["Weights ~45.1 GB (Q4_K_M), context ~1.5 GB at 16K tokens, runtime ~2.5 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 27.9 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/Qwen/Qwen3-Coder-Next-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3-Coder-Next"}},{"model":{"slug":"llama-3-3-70b","name":"Llama 3.3 70B","creator":"Meta"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":0,"confidence":"high","estimateGb":{"weights":39.6,"contextCache":5,"overhead":2.3,"total":46.9,"usableMemory":21.1,"offload":25.8},"utilization":2.22,"explanation":["Weights ~39.6 GB (Q4_K_M), context ~5.0 GB at 16K tokens, runtime ~2.3 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 25.8 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/unsloth/Llama-3.3-70B-Instruct-GGUF","huggingface":"https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct"}},{"model":{"slug":"qwen-agentworld-35b","name":"Qwen AgentWorld 35B-A3B","creator":"Alibaba Qwen"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":64,"confidence":"high","estimateGb":{"weights":20.6,"contextCache":0.31,"overhead":1.7,"total":22.6,"usableMemory":21.1,"offload":1.5},"utilization":1.072,"explanation":["Weights ~20.6 GB (Q4_K_M), context ~0.3 GB at 16K tokens, runtime ~1.7 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 1.5 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/unsloth/Qwen-AgentWorld-35B-A3B-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen-AgentWorld-35B-A3B"}},{"model":{"slug":"ornith-1-5-35b","name":"Ornith 1.5 35B-A3B","creator":"Ornith AI"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":65,"confidence":"high","estimateGb":{"weights":20.2,"contextCache":0.31,"overhead":1.7,"total":22.2,"usableMemory":21.1,"offload":1.1},"utilization":1.052,"explanation":["Weights ~20.2 GB (Q4_K_M), context ~0.3 GB at 16K tokens, runtime ~1.7 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 1.1 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF","huggingface":"https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B"}},{"model":{"slug":"kat-coder-v2-5","name":"KAT-Coder V2.5","creator":"Kwaipilot"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":67,"confidence":"high","estimateGb":{"weights":19.9,"contextCache":0.31,"overhead":1.7,"total":21.9,"usableMemory":21.1,"offload":0.8},"utilization":1.037,"explanation":["Weights ~19.9 GB (Q4_K_M), context ~0.3 GB at 16K tokens, runtime ~1.7 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 0.8 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/bartowski/Kwaipilot_KAT-Coder-V2.5-Dev-GGUF","huggingface":"https://huggingface.co/Kwaipilot/KAT-Coder-V2.5-Dev"}},{"model":{"slug":"qwen3-32b","name":"Qwen3 32B","creator":"Alibaba Qwen"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":58,"confidence":"high","estimateGb":{"weights":18.4,"contextCache":4,"overhead":1.7,"total":24.1,"usableMemory":21.1,"offload":2.9},"utilization":1.139,"explanation":["Weights ~18.4 GB (Q4_K_M), context ~4.0 GB at 16K tokens, runtime ~1.7 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 2.9 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/Qwen/Qwen3-32B-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3-32B"}},{"model":{"slug":"gemma-4-31b","name":"Gemma 4 31B","creator":"Google"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":66,"confidence":"high","estimateGb":{"weights":17.1,"contextCache":3.28,"overhead":1.6,"total":22,"usableMemory":21.1,"offload":0.9},"utilization":1.041,"explanation":["Weights ~17.1 GB (Q4_K_M), context ~3.3 GB at 16K tokens, runtime ~1.6 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 0.9 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/unsloth/gemma-4-31B-it-GGUF","huggingface":"https://huggingface.co/google/gemma-4-31B-it"}},{"model":{"slug":"granite-4-2-30b","name":"Granite 4.2 30B","creator":"IBM"},"quantization":"Q4_K_M","label":"OFFLOAD_REQUIRED","score":66,"confidence":"high","estimateGb":{"weights":16.5,"contextCache":4,"overhead":1.6,"total":22.1,"usableMemory":21.1,"offload":1},"utilization":1.046,"explanation":["Weights ~16.5 GB (Q4_K_M), context ~4.0 GB at 16K tokens, runtime ~1.6 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","About 1.0 GB would spill into system RAM (of roughly 64 GB). It runs, but noticeably slower than fully in VRAM."],"links":{"gguf":"https://huggingface.co/ibm-granite/granite-4.2-30b-GGUF","huggingface":"https://huggingface.co/ibm-granite/granite-4.2-30b"}},{"model":{"slug":"qwen3-8-2-4t-a95b","name":"Qwen3.8 2.4T-A95B","creator":"Alibaba Qwen"},"quantization":"IQ4_XS","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":1220.8,"contextCache":1.44,"overhead":37.7,"total":1260,"usableMemory":21.1,"offload":1238.8},"utilization":59.657,"explanation":["Weights ~1220.8 GB (IQ4_XS), context ~1.4 GB at 16K tokens, runtime ~37.7 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/unsloth/Qwen3.8-2.4T-A95B-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B"}},{"model":{"slug":"deepseek-v4-pro","name":"DeepSeek-V4-Pro","creator":"DeepSeek"},"quantization":"Q4_K","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":791.3,"contextCache":1.07,"overhead":24.8,"total":817.2,"usableMemory":21.1,"offload":796.1},"utilization":38.694,"explanation":["Weights ~791.3 GB (Q4_K), context ~1.1 GB at 16K tokens, runtime ~24.8 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/unsloth/DeepSeek-V4-Pro-0813-GGUF","huggingface":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813"}},{"model":{"slug":"glm-5-3","name":"GLM-5.3","creator":"Zhipu AI"},"quantization":"Q4_K","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":435.2,"contextCache":1.37,"overhead":14.2,"total":450.7,"usableMemory":21.1,"offload":429.6},"utilization":21.341,"explanation":["Weights ~435.2 GB (Q4_K), context ~1.4 GB at 16K tokens, runtime ~14.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/unsloth/GLM-5.3-GGUF","huggingface":"https://huggingface.co/zai-org/GLM-5.3"}},{"model":{"slug":"glm-5-2","name":"GLM-5.2","creator":"Zhipu AI"},"quantization":"Q4_K_M","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":433.8,"contextCache":1.37,"overhead":14.1,"total":449.3,"usableMemory":21.1,"offload":428.2},"utilization":21.273,"explanation":["Weights ~433.8 GB (Q4_K_M), context ~1.4 GB at 16K tokens, runtime ~14.1 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/unsloth/GLM-5.2-GGUF","huggingface":"https://huggingface.co/zai-org/GLM-5.2"}},{"model":{"slug":"deepseek-r1","name":"DeepSeek-R1","creator":"DeepSeek"},"quantization":"Q4_K_M","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":376.7,"contextCache":1.07,"overhead":12.4,"total":390.2,"usableMemory":21.1,"offload":369.1},"utilization":18.474,"explanation":["Weights ~376.7 GB (Q4_K_M), context ~1.1 GB at 16K tokens, runtime ~12.4 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/unsloth/DeepSeek-R1-GGUF","huggingface":"https://huggingface.co/deepseek-ai/DeepSeek-R1"}},{"model":{"slug":"ornith-1-5-397b","name":"Ornith 1.5 397B","creator":"Ornith AI"},"quantization":"Q4_K_M","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":227.5,"contextCache":0.47,"overhead":7.9,"total":235.9,"usableMemory":21.1,"offload":214.8},"utilization":11.169,"explanation":["Weights ~227.5 GB (Q4_K_M), context ~0.5 GB at 16K tokens, runtime ~7.9 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/ornith-ai/Ornith-1.5-397B-GGUF","huggingface":"https://huggingface.co/ornith-ai/Ornith-1.5-397B"}},{"model":{"slug":"glm-5-3-flash","name":"GLM-5.3-Flash","creator":"Zhipu AI"},"quantization":"Q4_K","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":186,"contextCache":0.17,"overhead":6.7,"total":192.9,"usableMemory":21.1,"offload":171.7},"utilization":9.131,"explanation":["Weights ~186.0 GB (Q4_K), context ~0.2 GB at 16K tokens, runtime ~6.7 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF","huggingface":"https://huggingface.co/zai-org/GLM-5.3-Flash"}},{"model":{"slug":"tencent-hy3","name":"Hy3","creator":"Tencent"},"quantization":"Q4_K_M","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":166.3,"contextCache":5,"overhead":6.1,"total":177.4,"usableMemory":21.1,"offload":156.3},"utilization":8.399,"explanation":["Weights ~166.3 GB (Q4_K_M), context ~5.0 GB at 16K tokens, runtime ~6.1 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/DevQuasar/tencent.Hy3-GGUF","huggingface":"https://huggingface.co/tencent/Hy3"}},{"model":{"slug":"deepseek-v4-flash","name":"DeepSeek-V4-Flash","creator":"DeepSeek"},"quantization":"Q4_K","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":144.4,"contextCache":0.76,"overhead":5.4,"total":150.6,"usableMemory":21.1,"offload":129.5},"utilization":7.13,"explanation":["Weights ~144.4 GB (Q4_K), context ~0.8 GB at 16K tokens, runtime ~5.4 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/unsloth/DeepSeek-V4-Flash-0731-GGUF","huggingface":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731"}},{"model":{"slug":"qwen3-8-flash-next","name":"Qwen3.8 Flash-Next","creator":"Alibaba Qwen"},"quantization":"Q4_K","label":"NOT_RECOMMENDED","score":0,"confidence":"high","estimateGb":{"weights":103.7,"contextCache":0.38,"overhead":4.2,"total":108.3,"usableMemory":21.1,"offload":87.2},"utilization":5.127,"explanation":["Weights ~103.7 GB (Q4_K), context ~0.4 GB at 16K tokens, runtime ~4.2 GB.","Your 24 GB leaves ~21.1 GB usable after a 12% safety margin.","Too far beyond your memory to run usefully, even with offloading."],"links":{"gguf":"https://huggingface.co/unsloth/Qwen3.8-Flash-Next-GGUF","huggingface":"https://huggingface.co/Qwen/Qwen3.8-Flash-Next"}},{"model":{"slug":"llama-3-1-8b","name":"Llama 3.1 8B","creator":"Meta"},"quantization":"Q4_K_M","label":"UNKNOWN","score":null,"confidence":"low","estimateGb":{"weights":4.6,"contextCache":null,"overhead":1.2,"total":null,"usableMemory":21.1,"offload":0},"utilization":null,"explanation":["We don't have this model's published attention shape yet, so we cannot estimate its context memory."],"links":{"gguf":"https://huggingface.co/unsloth/Llama-3.1-8B-Instruct-GGUF","huggingface":"https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct"}},{"model":{"slug":"llama-3-2-3b","name":"Llama 3.2 3B","creator":"Meta"},"quantization":"Q4_K_M","label":"UNKNOWN","score":null,"confidence":"low","estimateGb":{"weights":1.9,"contextCache":null,"overhead":1.2,"total":null,"usableMemory":21.1,"offload":0},"utilization":null,"explanation":["We don't have this model's published attention shape yet, so we cannot estimate its context memory."],"links":{"gguf":"https://huggingface.co/unsloth/Llama-3.2-3B-Instruct-GGUF","huggingface":"https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct"}}]}}