{"id":"p-h3-image","slug":"p-h3-image","name":"P-H3-Image","orgId":"pruna-ai","family":"p-h3-image","releaseDate":null,"availability":"ga","openWeights":true,"license":"MiniMax H3 Community License: excludes the EU, UK, Republic of Korea and US without a separate MiniMax license and requires \"Powered by MiniMax H3\"; Qwen-derived parts Apache 2.0, ComfyUI nodes MIT","paramsTotal":null,"paramsSource":null,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://www.pruna.ai/blog/oss-p-h3-image-launch","hfRepo":null,"modalities":["text","image"],"aliases":["P-H3-Image","prunaai/p-h3-image","p h3 image","ph3 image","p-h3","pruna h3 image","prunaai/p-h3-image-comfy"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-11-03","announced":"2026-11-03","blurb":"Pruna AI's open-weights text-to-image and image-editing model, rebuilt from the MiniMax-H3 video model: 8 sampling steps instead of 50, no classifier-free guidance, 2 to 16 MP output and edits with up to 8 input images. Weights on Hugging Face under the MiniMax H3 Community License, for Diffusers and ComfyUI.","api":"none","openWeights":"yes","params":{"text":"33.1B-parameter transformer per the card (66 GB in bf16), plus a 4B text encoder and a prompt upsampler based on Qwen3.5-9B; Pruna's blog describes the backbone as 20B after replacing H3's timestep block","source":"https://huggingface.co/PrunaAI/P-H3-Image"},"facts":[{"label":"Type","value":"Text-to-image and image editing (1 to 8 input images); single images only, no video or audio"},{"label":"Base model","value":"MiniMaxAI/MiniMax-H3 (a video model), fine-tuned and reworked: prompt upsampler, text encoder, diffusion schedule, transformer and decoder"},{"label":"Sampling","value":"8 steps (distilled from 50), no CFG or negative prompt"},{"label":"Resolution","value":"2 to 16 MP; the card recommends 8 MP for text-to-image and 4 or 8 MP for edits"},{"label":"Text encoder","value":"4B, distilled from H3's 32B encoder (9 GB vs 67 GB per the blog)"},{"label":"Prompt upsampler","value":"Included and on by default, based on Qwen3.5-9B"},{"label":"Weights","value":"Hugging Face PrunaAI/P-H3-Image (Diffusers ModularPipeline, about 100 GB in bf16) and PrunaAI/P-H3-Image-Comfy (int8 and 4-bit files for ComfyUI); not gated; a Playground on Hugging Face Spaces"},{"label":"Hardware","value":"About 101 GB GPU memory at 8 MP in bf16 with the upsampler (83 GB without), or 80 GB with offloading; the 4-bit ComfyUI build runs in about 21 GB"},{"label":"License","value":"MiniMax H3 Community License: excludes the EU, UK, Republic of Korea and US without a separate MiniMax license and requires \"Powered by MiniMax H3\"; Qwen-derived parts Apache 2.0, ComfyUI nodes MIT"},{"label":"API","value":"No hosted API stated in the launch materials"}],"claims":["Lab claim: 5x faster than MiniMax-H3 with 8 steps instead of 50: a 4 MP image in 2.6 s vs 12.4 s on one B200 (5.8 s vs 30.6 s at 8 MP).","Lab claim: 7x smaller than H3 at its smallest (a 20 GB 4-bit download vs 143 GB), running 8 MP images in about 21 GB of VRAM.","Lab claim: for most prompts, the 8-step model matches the 50-step original."],"sources":[{"kind":"lab","title":"P-H3-Image launch post (Pruna blog)","url":"https://www.pruna.ai/blog/oss-p-h3-image-launch","date":"2026-11-03"},{"kind":"lab","title":"Pruna AI announcement on X","url":"https://x.com/PrunaAI/status/2117663564487926043","date":"2026-11-03"},{"kind":"repo","title":"PrunaAI/P-H3-Image on Hugging Face","url":"https://huggingface.co/PrunaAI/P-H3-Image"},{"kind":"repo","title":"PrunaAI/P-H3-Image-Comfy on Hugging Face","url":"https://huggingface.co/PrunaAI/P-H3-Image-Comfy"}],"firstBenchmark":null},"organization":{"id":"pruna-ai","name":"Pruna AI","slug":"pruna-ai","country":null,"url":"https://www.pruna.ai","color":"#11110F","mark":"P"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Nov 3, 2026; no independent results yet.","lead":"As of Nov 22, 2026, P-H3-Image is pending benchmarks on isitsota: announced Nov 3, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}