Muse Spark 1.1

MarketScale reports that Meta's Muse Spark 1.1 entered the paid AI coding market at $1.25 per million input tokens, targeting developer teams and challenging Anthropic and OpenAI in the agentic coding sector. The model was launched on July 9 with a public developer preview API offering $20 in free credits for new accou Meta Superintelligence Labs, led by AI chief Alexandr Wang, trained Muse Spark 1.1 specifically to excel at coding-related tasks as a foundation for building effective AI agents. The model is designed to work with popular developer harnesses and third-party coding tools, directly responding to the surge in agentic AI a

NanoGPT published independent benchmark analysis on July 18, 2026 showing Meta's Muse Spark 1.1 scored 51 on the Artificial Analysis Intelligence Index, up from 43 for the original Muse Spark released three months earlier. The piece situates that score near GPT-5.4 at xhigh reasoning, GPT-5.6 Luna at max, and GLM-5.2 a The article frames Meta's first-party launch numbers as distinct from independent testing, noting Muse Spark 1.1 is text-and-image capable, supports tool calling, offers adjustable reasoning effort, and accepts up to one million input tokens. According to NanoGPT, the model's standout practical combination is strong be

Meta Superintelligence Labs released Muse Spark 1.1 on July 9, 2026, as a multimodal thinking model that builds on the original Muse Spark from April 2026. The model supports parallel subagents that handle smaller tasks and report back to a main agent, reducing completion time and compute overhead for complex workflows The model is accessible in thinking mode in the Meta AI app and will soon power experiences across Facebook, WhatsApp, and Instagram. For the first time, Meta is offering paid developer API access, a shift from its free model strategy designed to offset data center costs. Meta conducted safety testing against cybersecu

TechTimes reports that independent evaluator Artificial Analysis scored Muse Spark 1.1 at 51 on its Intelligence Index, an 8-point jump from the original Muse Spark's 43, with gains concentrated in scientific reasoning, coding, and knowledge. Per-task cost was estimated at roughly $0.26, beating GLM-5.2 at $0.37 and GP The piece notes that Artificial Analysis provided pre-release evaluation support to Meta, a relationship that tempers how the scores should be read, though the firm applies the same methodology across models with confidence intervals under ±1 percent. Output-token usage for the evaluation suite totaled 94 million for M

Fortune reports Meta released Muse Spark 1.1 on July 9, 2026, claiming it surpasses prior OpenAI, Anthropic, and Google models on coding and reasoning benchmarks. The launch is the second major Muse family release in the week, following the earlier rollout of Muse Image, and is overseen by chief AI officer Alexandr Wan Despite the broad claims, Fortune notes Muse Spark 1.1 still lags behind flagship entries including Anthropic's Mythos 5, Fable 5, and OpenAI's GPT-5.6 on at least one coding leaderboard. The piece also revisits past accusations that Meta manipulated test results for a prior AI release, which Meta has denied, while Zuc

Meta Superintelligence Labs released Muse Spark 1.1 on July 9, 2026, positioning it as a multimodal reasoning model optimized for agentic tasks, tool use, computer interaction, and coding, with availability through the new Meta Model API and in Meta AI. The model features a 1 million token context window for long-horiz Reported benchmark highlights include MCP Atlas (scaled tool use) at 88.1, JobBench at 54.7 (up from 17.0 for prior Muse Spark, beating Opus 4.8's 48.4 and GPT 5.5's 38.3), and OSWorld-Verified at 80.8, with the article also referencing Humanity's Last Exam. The model supports MCP for tool use, custom skills, structure

MarkTechPost reports that Meta Superintelligence Labs released Muse Spark 1.1 as a multimodal reasoning model designed for agentic tasks, available through the new Meta Model API. The release marks Meta's entry into the paid developer API market with a model optimized for coding, tool use, and complex multi-step workfl The coverage situates Muse Spark 1.1 within the competitive landscape against offerings from Anthropic and OpenAI, emphasizing its agentic capabilities and multimodal input handling. It underscores Meta's strategic shift toward monetized proprietary AI models alongside its ongoing open-source Llama efforts.

The New Stack reports that Meta has debuted Muse Spark 1.1, describing it as Meta's first paid AI model and noting its availability through an API aimed at developers building agentic applications. The article frames the launch as a significant commercial step for Meta, moving from open-source efforts into monetized mo Coverage emphasizes the model's positioning for coding and agentic workloads, aligning with Meta's broader strategy of competing directly with established paid offerings from OpenAI and Anthropic. The piece positions Muse Spark 1.1 as a meaningful entrant in the paid AI model market for technical users.

TechCrunch reports that Meta publicly launched Muse Spark 1.1 on July 9, 2026, as a multimodal AI model designed for agentic coding that competes directly with similar products from OpenAI and Anthropic. The model can perform multistep reasoning, manage digital workflows, and deploy new features in enterprise systems. The article notes pricing of $1.25 per million input tokens and $4.25 per million output tokens—roughly in line with Claude Haiku 4.5 and GPT-5.6 Luna—and highlights Meta's pitch around large agentic workloads, bug fixing, and large code migrations. CEO Mark Zuckerberg broke a three-year X posting hiatus to announce th

Meta released Muse Spark 1.1 on July 9, 2026, alongside a public preview of the Meta Model API, positioning it as a multimodal reasoning model built for agentic tasks with gains in tool use, computer use, coding, and multimodal understanding. The model is described as a significant upgrade from Muse Spark and updates t Rather than booting a pure benchmark leader, Muse Spark 1.1 is positioned as an agent foundation for AI founders, developer-tool companies, and automation startups targeting agents, tools, coding loops, browser work, multimodal workflows, and long-running tasks. The key framing is whether it can complete useful work at

The BenchLM tracker for Muse Spark 1.1 reports a 65.9/100 capability score, with verified category rankings of 22nd in Agentic (57.7, 80th percentile across 14 benchmarks), 23rd in Coding (56.3, 84th percentile across 4 benchmarks), 15th in Knowledge (68.2, 91st percentile across 5 benchmarks), and non-ranked but verif The page flags Muse Spark 1.1 as "Superseded" by Muse Spark 1.2, with the data snapshot dated September 24, 2026, indicating the variant has aged relative to newer Muse Spark releases. Despite the supersession status, the tracker still offers direct benchmark comparisons against the broader ranked and priced model cata