Meta has made a strategic pivot by unveiling Muse Spark 1.1, its first commercial language model. Unlike its previous strategy focused exclusively on open-source solutions like Llama, the company is entering the segment of automated code writing and autonomous AI agents, where Anthropic and OpenAI currently reign supreme. And it is entering aggressively.
Price War: Meta Strikes at Competitors' Bottom Line
Muse Spark's key advantage is its price. The cost of processing 1 million input tokens is just $1.25, while generating 1 million output tokens costs $4.25. This is radically lower than competitors' rates. For comparison: Anthropic Sonnet 5 charges $2 for input and $10 for output; advanced models like Opus 4.8 and GPT-5.5 charge $5 for input and $25-$30 for output, respectively.
The difference is staggering: Muse Spark's input tokens are 37% cheaper than Sonnet 5 and 75% cheaper than Opus/GPT-5.5. For output tokens, the gap is even wider — up to 86% compared to GPT-5.5. Even after Sonnet 5's promotional rate ends on August 31, the standard pricing of $3/$15 still falls short of Meta's offer.
Does Muse Spark Have a Chance to Outpace Competitors?
Early feedback from developers who have gained access to the platform is encouraging. Reports indicate that Muse Spark demonstrates performance comparable to GPT-5.5 and Claude Opus 4.8 in complex tests for autonomous task execution. However, Meta has not yet provided independent benchmarks, leaving room for skepticism.
Access to the model is strictly limited — only to specialists in the US after application approval. The AI is not yet available on popular aggregators like OpenRouter. Nevertheless, the low entry barrier could eventually attract a significant portion of the IT community, especially cost-sensitive startups and small-to-medium businesses.
Cryptalist Expert Analysis: Muse Spark is not just another model, but a clear signal of Meta's intention to undercut prices until competitors are forced to reduce their margins. In the short term, this benefits consumers, but in the long term, it could lead to market consolidation around the largest players capable of subsidizing losses. The question remains whether Muse Spark's quality can match its aggressive pricing strategy, or if we are getting a classic case of "cheap but rough."