{"id":"pan-2","slug":"pan-2","name":"PAN 2","orgId":"mbzuai-ifm","family":"pan-2","releaseDate":null,"availability":"announced","openWeights":true,"license":null,"paramsTotal":null,"paramsSource":null,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://x.com/IFM_AI/status/2120199814696828942","hfRepo":null,"modalities":["world-model","video"],"aliases":["PAN 2","pan-2","PAN2","pan 2","IFM PAN 2","IFM/PAN2","ifm-ai/PAN2"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-11-10","announced":"2026-11-10","blurb":"IFM (MBZUAI)'s open foundation world model for physical intelligence. A multimodal LLM (the conceptual reasoning backbone) writes the next world state as a structured, time-grounded description of events, actions and camera motion; a 29B latent diffusion model trained from scratch (the perceptual reasoning backbone) turns that description into a video of up to 20 seconds. Released November 10, 2026 with public weights and training and inference code under Apache License 2.0. Supported in Hugging Face Diffusers and vLLM.","api":"none","openWeights":"yes","params":{"text":"29B-parameter MMDiT perceptual reasoning backbone, trained from scratch. Conceptual reasoning is a pre-trained multimodal LLM (Kimi K3 is used in the technical report unless stated otherwise). Frozen text encoder is Qwen3.5-9B; VAE is Wan 2.2, both used unmodified from their authors","source":"IFM PAN 2 Hugging Face model card and GitHub README"},"facts":[{"label":"Type","value":"General-purpose world model with Generative Latent Prediction (GLP): conceptual reasoning by a multimodal LLM plus perceptual reasoning by a 29B latent diffusion model (IFM page, model card and GitHub)"},{"label":"Output","value":"Video via the Wan 2.2 VAE decoder, 704×1248 at 24 fps, up to 20 seconds; text-to-video and image-to-video from one diffusion checkpoint (model card)"},{"label":"Weights","value":"Public, ungated Hugging Face repo IFM/PAN2 (BF16 diffusion backbone). The Qwen3.5-9B text encoder and Wan 2.2 VAE are downloaded from their own repositories"},{"label":"License","value":"PAN 2 code and weights Apache License 2.0 (IFM announcement, model card and GitHub). Third-party Wan 2.2, Qwen3.5-9B and Diffusers components keep their own licenses"},{"label":"Code","value":"https://github.com/ifm-ai/PAN2 (training and inference code; also Diffusers, vLLM and FastVideo)"},{"label":"API","value":"No public cloud API announced; run locally from the released weights and code"},{"label":"Related","value":"Separate from earlier PAN, from Jais 3 (same lab), and from the frozen Wan 2.2 VAE and Qwen3.5-9B encoder used as components"}],"claims":["Lab claim: on 100 egocentric everyday multi-step tasks posed as I2V and T2V (200 prompts, 456 steps), with Kimi K3 as the conceptual reasoner, PAN 2 records 75.10 ± 1.51% step completion and 34.00 ± 3.61% full-plan completion, mean ± SD over 3 seeds (IFM technical report figure on the Hugging Face model card).","Lab claim: six-capability overall accuracy of 84.48 ± 0.08% (physical consistency, causal coherence, entity persistence, constraint satisfaction, viewpoint control and external dynamism; 300 single-action test cases; IFM technical report figure on the Hugging Face model card).","Lab claim: Physics-IQ Verified (I2V) 49.38 ± 0.64 in the public leaderboard setting, with Claude Opus 5.5 as the conceptual reasoner, mean ± SD over seeds 123–126 (IFM technical report figure on the Hugging Face model card)."],"sources":[{"kind":"lab","title":"IFM on X: Introducing PAN 2","url":"https://x.com/IFM_AI/status/2120199814696828942","date":"2026-11-10"},{"kind":"lab","title":"IFM on X: PAN 2 weights, training and inference code under Apache 2.0","url":"https://x.com/IFM_AI/status/2120199815724507276","date":"2026-11-10"},{"kind":"lab","title":"IFM PAN 2 page","url":"https://ifm.ai/pan/","date":"2026-11-10"},{"kind":"repo","title":"IFM/PAN2 on Hugging Face","url":"https://huggingface.co/IFM/PAN2"},{"kind":"repo","title":"ifm-ai/PAN2 on GitHub","url":"https://github.com/ifm-ai/PAN2"}],"firstBenchmark":null},"organization":{"id":"mbzuai-ifm","name":"IFM (MBZUAI) with Inception42","slug":"mbzuai-ifm","country":null,"url":"https://ifm.ai","color":"#11110F","mark":"I"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Nov 10, 2026; no independent results yet.","lead":"As of Nov 12, 2026, PAN 2 is pending benchmarks on isitsota: announced Nov 10, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}