8.4K views · 233 likes · 28 reposts (at collection)
MUSE WONDER this is the culmination of over a year of research and infrastructure work in msl since we originally shipped muse spark in april, we have continued to scale our models across pre-training and RL, now reaching the frontier 🏆 see our model's improvements on the artificial analysis index:
- Who
- Meta
- What
- announced improvements to the Muse Wonder model on the Artificial Analysis index
What happened
MUSE WONDER this is the culmination of over a year of research and infrastructure work in msl since we originally shipped muse spark in april, we have continued to scale our models across pre-training and RL, now reaching the frontier 🏆 see our model's improvements on the artificial analysis index
4 reports
The story so far
Meta released Muse Wonder through the Meta Model API and Muse Code and Muse Wonder Full for Muse Code subscribers, while Alexandr Wang said the team was addressing a bug in how both models counted against subscribers’ usage limits. Wang promised resets once the bug was fixed; his subsequent update said issues remained. TechCrunch reports that Wonder’s API pricing is $4.00 per million input tokens and $20.00 per million output tokens. Artificial Analysis reports that Wonder and Wonder Full scored 52 and 56 on its Intelligence Index v4.3 at maximum reasoning effort, compared with 48 for Muse Spark 1.3 at the same setting. Full improved across all ten evaluations, with its largest gains in agentic evaluations. Both models support a one-million-token context window. Meta says its internal benchmarks show performance competitive with GPT-6 Astra, Claude Opus 5.5 and Claude Fable 5.1, while The Decoder reports weaker factual knowledge than leading competitors despite stronger coding and agent results. Statements about access through Meta’s personal agent, Muse, differ: Meta’s announcement says Wonder will power it soon, starting with subscribers, while Baoyu reports that Zuckerberg said subscribers could use it at launch. In his latest comments, Wang acknowledged there was still much to improve in response to a Wonder Full review that praised its writing and orchestration but criticized pricing and usage limits.
Keep reading
Latest daily · Issue 24NVIDIA and AI4Bharat Introduce Nemotron-pretraining-Indic-v1 and an Open Data Generation PipelineTrending now
- 1Meta Releases Muse Spark Open With Mac and Cloud Support, Topping US Open-Weight Index
- 2Google Launches Gemini 4 Argon Globally, Expands Paid Access as Benchmark Results Vary
- 3OpenAI Updates GPT-6 Astra Pro in ChatGPT With Parallel Agents and Faster Responses
- 4Meta’s New Business Terms Raise Dispute Over Website Crawling and AI Training Rights
Tomorrow at 08:00, a briefing like this arrives in your inbox.
The same kind of briefing as the one above: a few stories worth reading. Free. Mail starts after you confirm, and every letter has a one-click unsubscribe. Read a past issue · 2026-10-17
