Muse Wonder
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 18
- StarSkirmish Bench v0.2: GPT-6 Astra tops LLM StarCraft bot ranking
StarSkirmish Bench v0.2 gives each large model one hour to write a Protoss bot in C++ for StarCraft: Brood War, then scores it by win rate against human-written and demo bots. GPT-6 Astra leads with 57, followed by GPT-6.1 Sol at 46, with Gemini 4 Argon and Claude Opus 5.5 tied at 39, Claude Fable 5.1 at 37, then Claude Sonnet 5.5 (26), Muse Wonder Full and Claude Haiku 5.5 (21) and GPT-6 Luna (13). Even the best model, GPT-6 Astra, wins only about 7% of games against top human-written bots Stardust, PurpleWave and BananaBrain, so the benchmark is far from saturated; practicing against strong opponents is the strongest predictor of a model's score, with Claude Fable 5.1 a clear outlier.
Hacker News · AI（100+ 分） · 🔥 20 - AI safety worries mount as the industry keeps racing ahead
This week's enterprise AI roundup reports that after a run of "rogue agent" incidents, more government officials are calling for regulation, several states are suing OpenAI, about 400 scientists urged global AI oversight, and Microsoft's president said AI is less regulated than a carton of milk. On the money side, Anthropic told investors revenue grew 45% sequentially to $16.8B in Q3, a slowdown that could pull its planned IPO valuation from $2T toward $1.5T, while CoreWeave's debt load weighed on neocloud and supplier shares. On models and products, Meta released the 816B-parameter open-weight Muse Spark Open for coding and agent work plus Muse Wonder for personal and professional agents, Nvidia expanded its Nemotron family, and Grok Voice 3.0, Gemini 4 Argon and Microsoft's Maia 300 also arrived, alongside news that Character.AI filed for bankruptcy.
SiliconANGLE AI · 🔥 8 - Meta Muse's rise: can personal AI agents scale beyond walled gardens?
An analysis republished by 36Kr (originally from Haike Caijing) says Meta's Muse personal AI agent, launched Sept. 8, topped the U.S. iOS free app chart for multiple days, with Sensor Tower reporting more than 730,000 downloads in five days and over 5 million cumulative by Sept. 30; Meta followed with the long-horizon Muse Wonder model on Oct. 12, a day before Google's Gemini 4 Argon. The piece says Muse chains together booking, price comparison and email tasks through users' email and Google accounts, but Amazon has asked Meta to remove it from the service, while Shopify opened checkout to AI agents via a universal commerce protocol; in China, agents face app-island barriers, and Doubao's phone assistant listed 151 apps—including Taobao, Alipay and WeChat—that refuse agent operation. It also cites Muse pricing (roughly 100 million free tokens a week, plus $20 and $100 monthly tiers), a reported Mac client zero-day, a six-bank risk warning, and a Toronto blogger who said Muse accepted a low offer and shared his home address.
36氪 人工智能 · 🔥 15 - Claude Fable 5.5 surfaces in Azure catalog with 1M context, hinting at imminent launch
A 36Kr repost of a New Intelligence Era (新智元) report says a model entry named azure_ai/claude-fable-5-5 appeared in Microsoft Azure's East US 2 catalog listed as Generally Available, while the open-source project LiteLLM synced a matching configuration, which observers read as a signal that Anthropic's next model is close to launch. The leaked specifications describe a 1M-token context window, up to 128K output tokens, and API pricing of $10 per million input tokens and $50 per million output tokens. The report cautions that a GA listing in Azure's catalog does not equal an official announcement, and that LiteLLM following cloud catalogs is routine, so it does not count as a second independent source. It also cites an Anthropic engineering blog saying internal code merges grew from about 2,500 per day in May to more than 30,000 per day in September, and speculates from that figure that Fable 5.5 may have been produced through AI self-iteration — a claim offered by the author and commenters, not confirmed by Anthropic.
36氪 人工智能 · 🔥 7 - Analysis: Meta's AI edge is its ad machine, not its models
Meta's Muse agent, released in October 2026, reached 73% early user retention — which the article says is close to double ChatGPT's figure at a comparable stage — alongside the flagship Muse Wonder and Muse Wonder Full models. But an analysis republished by 36Kr argues Meta's real advantage is how its mature ad business funds and absorbs AI: Q2 2026 ad revenue was $59.363 billion, 97.6% of total revenue and up 27% year over year, while 2026 capital spending is projected at $115–135 billion. The piece notes that Muse Image has been folded into the Advantage+ ad platform, that 9 million small businesses had used at least one AI ad-creation tool as of July 2026, and that Advantage+ end-to-end campaigns passed $75 billion in annualized revenue; it compares this with Tencent (marketing services revenue up 20% to 38.2 billion yuan, with Yuanbao embedded in WeChat) and ByteDance (Doubao linked to Douyin e-commerce, Seedance 2.5 in TikTok Symphony). It cites a September 2026 Morgan Stanley survey in which 72% of advertisers reported unchanged total budgets, suggesting AI is mainly reshuffling existing spend before 2030.
36氪 人工智能 · 🔥 6 - Analyst shares first take on Meta's Muse Wonder and Mistral's Le Chonk
Arena's Model Insights Analyst Dawid Galarowicz shared his initial take on Meta's Muse Wonder and Mistral's Le Chonk, along with two AI trends he says are worth watching next, in an X post linking to a video. The post itself gives no specifications, benchmarks or release details for either model; the commentary reflects the analyst's own first impressions.
arena (X) · 🔥 4 - Anthropic tells investors Q3 revenue hit $16.8B ahead of IPO
Anthropic told prospective investors at a presentation at its San Francisco headquarters that third-quarter revenue reached $16.8 billion, up from $11.5 billion in the second quarter — a roughly 45% sequential jump, according to Bloomberg, The Information and other outlets citing people familiar with the matter. The company also said it was profitable on an adjusted basis in the quarter, with an 85% adjusted gross margin, an adjusted operating margin above 11% and positive free cash flow. Sources said Anthropic expects to begin its IPO roadshow as early as the week of Nov. 9 and could list before Thanksgiving, though discussions are ongoing and timing may change.
Bloomberg Technology · 🔥 67 - Meta's Muse Wonder (Max) debuts at #12 in Agent Arena, topping Muse Spark 1.3
Arena.ai published results from more than 5,000 real-world agentic sessions showing Meta's Muse Wonder (Max) entering the Agent Arena at #12 with a +6.4% net improvement, 2.8 points and four places above Muse Spark 1.3 (Max) at #16 with +3.6%. The gain comes at a higher price: a $1.13 median cost per task versus $0.54 for Spark 1.3, about 2.1x. Arena describes Muse Wonder (Max) as cost-efficient relative to nearby models — within 1 percentage point of #10 GPT-5.6 Sol (xHigh) at 25% lower cost, and within 2 points of #9 Claude Fable 5 (High) at 67% lower cost. By category it leads Muse Spark 1.3 in Chat (#12, +6.5% vs #15, +3.3%), Work (#13, +6.8% vs #18, +3.7%) and Code (#15, +5.2% vs #18, +4.2%), but in Code Arena: WebDev it placed #15 with 1651 points, slightly below Muse Spark 1.3's #12 and 1659 points, while landing #2 in Text Arena with 1513 points.
arena (X) · 🔥 3 - fal tutorial: build an AI filmmaking chat agent with Muse Wonder in T3 Code
fal shared a tutorial on X showing how to build an AI filmmaking chat agent using Muse Wonder Full inside T3 Code, with a link to a YouTube video. The post does not describe the model's capabilities, pricing or the steps in the tutorial, beyond naming filmmaking as the use case. Everything here comes from fal's own promotional post, with no other sources or independent verification.
fal (X) · 🔥 0 - Meta's Muse Wonder tops coding benchmarks but flunks a lava lamp test
Meta released Muse Wonder and Muse Wonder Full on Oct 12, and official figures put them in the top tier on several benchmarks: Artificial Analysis Intelligence Index scores of 52 and 56, between Claude Opus 5.5 (58) and GPT-6 Astra (53), plus 57.5 on SWE-Marathon for the Full version versus 46.3 for Claude Opus 5.5. But in an open-ended real-world test by evaluator BridgeMind, the model failed to build a lava lamp with a transparent glass shell, moving lava and realistic lighting, producing what the article calls an ordinary-looking desk lamp; BridgeMind accused it of "benchmaxxing," and Elon Musk replied "Hmm" on X. The article also notes that SWE-Marathon actually grades a full coding-agent setup, including Codex and Claude Code, rather than a bare model, and that Meta's focus has shifted from the Llama open ecosystem toward a personal assistant built around long-term memory and sustained task execution.
虎嗅 AI · 🔥 0 - Meta releases Muse Wonder frontier model, Zuckerberg details $100B-plus bet
Meta released the Muse Wonder frontier model on Oct. 13, a much larger successor to Muse Spark that Meta says leads on personal and professional agent tasks and in finance and legal evaluations, with strong coding results on DeepSWE, SWE-Marathon and Terminal-Bench 4, plus an improved "active memory" system tested on its internal Memory Bench. The model is available through Meta's model API and OpenRouter, with Muse Wonder Full reserved for Muse Code subscribers; Cursor also added it, reporting a 46.1% CursorBench score at default effort and pricing of $4 per million input and $20 per million output tokens. In a podcast interview, Zuckerberg and Meta Superintelligence Labs chief AI officer Alexandr Wang said running a 1 GW cluster costs about $15 billion a year with $50–60 billion in capex, that training a frontier model now runs into tens of billions of dollars, and that Meta committed $100–200 billion before seeing any model results.
智东西 · 🔥 7 - Personal agent wave: Qwen Go opens invite testing as Meta ships Muse Wonder
At its Yunqi conference in late September, Alibaba said Qwen would accelerate its push into personal agents; on October 13 it unveiled the personal assistant "Qwen Go" with limited invite testing in the Qwen app. ByteDance's Doubao is reported to be testing a personal agent project codenamed Spell and merging it with existing products, while Tencent has TX Handy Bot and Yuanbao Assistant; Meta's Muse is credited with igniting the wave, and OpenAI followed with dots. On October 12 Meta released the Muse Wonder series (Muse Wonder and Muse Wonder Full), centered on Active Memory and Active Focus, described as ranking just behind Opus 5.5 on an AA leaderboard, with internal MemoryBench and FocusBench tests for long-dialogue recall and sustained progress. Qwen's team also published QwenGyre, an online reinforcement learning framework for very long-horizon agent tasks, submitted to arXiv on September 27; it reports Qwen 3.8 2.4T lifting NL2RepoBench pass rate from about 52.5% to 58.5% after 48 RL steps and cutting training time roughly 44% versus an Async baseline at similar scores.
36氪 人工智能 · 🔥 0 - Prime Intellect acquires Moondream, adds Moondream 3.1 and Photon to its inference stack
Prime Intellect announced it has acquired Moondream, with founders Vik Korrapati and Jay Allen joining to lead and scale inference and to accelerate multimodal work spanning vision and computer use. Moondream 3.1 is now available through Prime Inference powered by Moondream's Photon engine, and Photon plus Moondream's GPU kernel and megakernel compiler work are being integrated into Prime's stack. Clef and Clef Flash are also live through Prime Inference and OpenRouter, with Clef priced at parity with Jev and Clef Flash at half the price; existing Moondream Cloud customers will receive platform credits in the coming days.
Prime Intellect Blog · 🔥 0 - Argon undercuts frontier rivals as cost per task varies 3.8x
A Tomasz Tunguz analysis cites Artificial Analysis data (retrieved October 2026) showing five frontier models sitting within about five intelligence points of each other while cost per task differs 3.8x, with Google's Argon at the bottom at $1.99 per task — 74% below the priciest model, Fable 5.1, and 67%, 56% and 39% below Claude Opus 5.5, Muse Wonder and Astra respectively, all measured at maximum reasoning effort. The author attributes the gap to custom silicon, noting Google's TPU effort since 2013 and estimating it could serve tokens roughly 44% cheaper than peers, while Microsoft's Maia 300 (introduced October 2026, claimed 40% more tokens per watt than leading GPU-based systems) and OpenAI's first inference chip, Jalapeño (1.5x–1.9x more throughput per watt and 1.7x–3.6x lower end-to-end latency against GB200/GB300 on SemiAnalysis's InferenceX suite), follow the same playbook. The piece flags caveats: Maia 300's figure is company-reported with no independent benchmark, SemiAnalysis called the Blackwell comparison “somewhat incomplete and unfair,” and the 40–70% margin range is the author's own estimate.
Tomasz Tunguz · 🔥 0 - How Muse Wonder pulled Meta back to the AI frontier
A licensed analysis republished by TMTPost says Meta Superintelligent Labs released Muse Wonder and Muse Wonder Full on Oct 12, 2026, scoring 52 and 56 on Artificial Analysis's intelligence index — the Full version level with Anthropic's Claude Sonnet 5.5 and ahead of GPT-6 Astra and Gemini 4 Argon. It credits Scale AI founder Alexandr Wang, brought into Meta as chief AI officer in June 2025 through a $14.3B investment, with betting on personal agents built around Active Memory and Active Focus instead of head-on competition with OpenAI and Anthropic on coding and abstract reasoning. The piece adds that Anthropic's Claude Opus 5.5 still leads the index at 58, that GPT-6.1 Sol retains a cost-efficiency edge, and that Muse Wonder Full is currently limited to Muse Code subscribers with enterprise availability unclear.
钛媒体 · 🔥 0 - First hands-on comparison of Meta's Muse Wonder
Arena shared a video in which reviewer @petergostev puts Meta's Muse Wonder (Max) through one-shot 3D scenes, animated HTML and front-end design tasks. He compares it against Astra, Opus 5.5, GPT-6.1 Sol, Claude Sonnet 5.5, Fable, Kimi and Mistral, and includes the cost of every generation. Arena says its scores for Muse Wonder are coming very soon, so this is currently only a first-impressions look.
arena (X) · 🔥 0 - Meta unveils Muse Wonder, claiming state-of-the-art performance
Meta released a new model called Muse Wonder, which it claims can "deliver state-of-the-art performance" on certain tasks, marking what The Information frames as a comeback after an AI slump and a roughly 14-month hiring binge. The article notes that nearly every major lab is making similar claims: Google introduced Gemini 4 Argon two weeks earlier, saying it "delivers frontier performance in complex workflows"; OpenAI announced GPT-6 Astra in early September, calling it "state of the art on computer use, browsing, software engineering, cybersecurity, science and professional work"; and Anthropic launched Claude Fable 5.1 and Mythos 5.1 in September, describing them as "the world's most advanced models for coding and knowledge work." With everyone claiming the frontier, the piece asks whether users will actually care.
The Information · 🔥 0 - Meta releases Muse Wonder and Muse Wonder Full, claiming frontier agentic performance
Meta released two new models, Muse Wonder and Muse Wonder Full, on October 12, saying they deliver state-of-the-art performance on personal and professional agentic tasks, with Muse Wonder Full described as having frontier capabilities. Muse Wonder is available to Muse Code subscribers starting today and can also be accessed through the Meta Model API; Meta says it eventually plans to power its Muse personal agent with Wonder, starting with subscribers, but gave no specifics or timeline. Meta cites internal benchmark testing to claim the models rival leading offerings such as OpenAI's GPT-6 Astra and Anthropic's Claude Opus 5.5, and says its largest version beats Claude Fable 5.1 and ranks just below Claude Opus 5.5 on the Artificial Analysis Intelligence Index. Meta did not disclose how the models were built or trained, and the comparisons rest on its own testing.
36氪 人工智能 · 🔥 0
Experience and discussion from the community
Share my Muse Wonder experienceAsk about Muse Wonder
Nobody has shared their experience with Muse Wonder yet.