Muse Wonder Full is the most capable model in Meta's Muse Wonder series, built for long-running professional agents and ultra-long-horizon software engineering. It shares Muse Wonder's Active Memory and Active Focus capabilities and generally scores higher on Meta's agentic and coding evaluations.
Provider-specific identifiers, limits, and listed prices per million tokens. Every row links back to the provider's own documentation.
The model record exists, but no source-linked provider offer is available yet.
Submit a sourceRecorded from the source catalog and provider listings.
meta/muse-wonder-fullEvery result stays attached to its source, version, metric and harness. Scores from different versions are never merged, and the profile below plots each benchmark against its own population rather than on a shared scale.
Each strip shows every published score for that benchmark, with this model marked. The lighter shape behind the ticks is the density of results, and the dashed line is the median.
Field-level changes detected between successful source imports.
This record has not changed within the retained import history.
Every figure on this page traces back to one of these records.
Fetch the complete source-linked model record. No key, no account, no rate-limited tier.
GET https://modelbench.lol/api/v1/models/meta/muse-wonder-fullcurl "https://modelbench.lol/api/v1/models/meta/muse-wonder-full"Answered directly from the stored record — nothing here is generated beyond the catalog's own fields.
Muse Wonder Full is the most capable model in Meta's Muse Wonder series, built for long-running professional agents and ultra-long-horizon software engineering. It shares Muse Wonder's Active Memory and Active Focus capabilities and generally scores higher on Meta's agentic and coding evaluations. It is published by Meta and catalogued here from Models.dev.
Muse Wonder Full accepts up to 1.05M tokens of context.
Provider catalogs list support for image input.
No. This model is served through hosted APIs only.
The catalog records a release date of 2026-10-12, last verified Oct 24, 2026.
Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification. Similar to previous versions, it can be used to classify content in both LLM...
Meta: Llama 4 MaverickLlama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...
Meta: Llama 4 ScoutLlama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input...
Meta: Llama 3.3 70B Instruct (free)The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...