Meta's Muse Spark 1.1 takes aim at OpenAI and Anthropic with stronger agentics, multistep reasoning, and a price advantage that could shift enterprise AI spending.
Alpha Score of 62 reflects moderate overall profile with weak momentum, strong value, strong quality, moderate sentiment.
Meta released Muse Spark 1.1 Wednesday, an updated AI model with improvements to coding, multimodal reasoning, and agentic workflows. The company also announced pricing changes that undercut several competing models on a per-token basis.
The timing is deliberate. Mark Zuckerberg said in a statement that the update is meant to push the AI race onto tougher ground, where capability and cost both matter. The model can handle more complex code generation, parse images alongside text, and chain multiple reasoning steps, according to Meta's release.
Pricing is the sharper edge. Muse Spark 1.1 costs less per million input tokens than comparable models from OpenAI and Anthropic. The gap widens on higher-volume use cases, which could shift decision-making for startups and enterprise teams that have watched costs climb over the past year.
Meta is also opening more of the model's weights to developers, a move that extends the company's strategy of courting the open-source AI community. The approach has grown Meta's user base among independent researchers and smaller AI labs, though it has not yet produced the enterprise revenue that rivals generate through API sales alone.
Shares of Meta Platforms fell 1.75% to $592.59 Wednesday. The dip came alongside a broader tech sector pullback; analysts at several firms said the stock price move had more to do with positioning than news about Muse Spark's release.
The agentic features in Spark 1.1 may matter more over time. The model can plan, execute corrections, and hand off subtasks without restarting, a set of behaviors that reduces developer overhead on long-running AI workflows. Few competitors have released similarly granular agent functionality into general availability.
Meta's Alpha Score stands at 62 out of 100, a moderate rating that reflects the company's strong cash flows and dominant ad business alongside the higher spending required to maintain AI infrastructure. The spending demands are real. Capital expenditure this year is on track to exceed Wall Street initial estimates, a gap that some shareholders have flagged.
Zuckerberg framed the new release as a necessary offensive move. "We are not building this to match what exists," he said in the statement. "We are building to change what the next generation of AI looks like."
Prepared with AlphaScala research tooling and grounded in primary market data: live prices, fundamentals, SEC filings, hedge-fund holdings, and insider activity. Each story is checked against AlphaScala publishing rules before release. Educational coverage, not personalized advice.