{"as_of":"2026-10-12","license":"CC-BY-4.0","license_url":"https://creativecommons.org/licenses/by/4.0/","cite_as":"The Local Bench (thelocalbench.com), data as of 2026-10-12, CC BY 4.0","source_url":"https://thelocalbench.com","model":{"id":"qwen3-32b","name":"Qwen3 32B","org":"Alibaba Qwen","task":"general","hf_repo":"Qwen/Qwen3-32B","gguf_repo":"Qwen/Qwen3-32B-GGUF","license":{"value":"Apache-2.0","flag":"verified","source":"https://www.apache.org/licenses/LICENSE-2.0"},"datasheet":"TLB-M-0008","rev":"2026.10","as_of":"2026-10-12","summary":"The largest dense Qwen3. Q4_K_M weights are 19.8 GB, which is tight on a 24 GB card once context grows.","arch":{"params_total":{"value":32.76,"flag":"verified","unit":"B","source":"https://huggingface.co/api/models/Qwen/Qwen3-32B","note":"safetensors.total from the HF API."},"params_active":{"value":32.76,"flag":"verified","unit":"B","source":"https://huggingface.co/api/models/Qwen/Qwen3-32B","note":"Dense: all parameters active."},"layers":{"value":64,"flag":"verified","source":"https://huggingface.co/Qwen/Qwen3-32B/blob/main/config.json"},"kv_heads":{"value":8,"flag":"verified","source":"https://huggingface.co/Qwen/Qwen3-32B/blob/main/config.json"},"q_heads":{"value":64,"flag":"verified","source":"https://huggingface.co/Qwen/Qwen3-32B/blob/main/config.json"},"head_dim":{"value":128,"flag":"verified","source":"https://huggingface.co/Qwen/Qwen3-32B/blob/main/config.json"},"hidden_size":{"value":5120,"flag":"verified","source":"https://huggingface.co/Qwen/Qwen3-32B/blob/main/config.json"},"context":{"value":32768,"flag":"sourced","unit":"tokens","source":"https://huggingface.co/Qwen/Qwen3-32B","note":"config.json lists max_position_embeddings 40,960; the model card gives 32,768 native and 131,072 with YaRN."}},"quants":{"Q3_K_M":{"bpw":3.91,"file_gb":{"value":16.01,"flag":"estimate","unit":"GB","note":"params_total × 3.91 bpw ÷ 8 (brief §7.3 fallback); no GGUF of this quant in Qwen/Qwen3-32B-GGUF."},"kld":{"value":0.071,"flag":"sample","note":"Placeholder until KLD runs (brief §4.2)."}},"Q4_K_M":{"bpw":4.85,"file_gb":{"value":19.76,"flag":"verified","unit":"GB","source":"https://huggingface.co/Qwen/Qwen3-32B-GGUF/tree/main","note":"Sum of GGUF file sizes (Q4_K_M) from the HF API tree listing."},"kld":{"value":0.017,"flag":"sample","note":"Placeholder until KLD runs (brief §4.2)."}},"Q5_K_M":{"bpw":5.69,"file_gb":{"value":23.21,"flag":"verified","unit":"GB","source":"https://huggingface.co/Qwen/Qwen3-32B-GGUF/tree/main","note":"Sum of GGUF file sizes (Q5_K_M) from the HF API tree listing."},"kld":{"value":0.008,"flag":"sample","note":"Placeholder until KLD runs (brief §4.2)."}},"Q6_K":{"bpw":6.56,"file_gb":{"value":26.88,"flag":"verified","unit":"GB","source":"https://huggingface.co/Qwen/Qwen3-32B-GGUF/tree/main","note":"Sum of GGUF file sizes (Q6_K) from the HF API tree listing."},"kld":{"value":0.004,"flag":"sample","note":"Placeholder until KLD runs (brief §4.2)."}},"Q8_0":{"bpw":8.5,"file_gb":{"value":34.82,"flag":"verified","unit":"GB","source":"https://huggingface.co/Qwen/Qwen3-32B-GGUF/tree/main","note":"Sum of GGUF file sizes (Q8_0) from the HF API tree listing."},"kld":{"value":0.001,"flag":"sample","note":"Placeholder until KLD runs (brief §4.2)."}},"F16":{"bpw":16,"file_gb":{"value":65.52,"flag":"estimate","unit":"GB","note":"params_total × 16.0 bpw ÷ 8 (brief §7.3 fallback); no GGUF of this quant in Qwen/Qwen3-32B-GGUF."}}},"quality":{"arena_text":{"value":1340.1,"flag":"sourced","source":"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset","note":"LMArena text, overall, entry 'qwen3-32b': 1340.1 (95% CI 1330.8–1349.4, 3,926 votes), leaderboard published 2026-10-08. CC-BY-4.0."}}},"runs":[]}