Meta's New Model is Cheap, But Only on the API
hero88645 · reddit · 2026-07-15
This post breaks down the pricing and positioning of **Meta's new model, Muse Spark 1.1**. It is Meta's first truly API-billed model, priced at **$1.25 / $4.25 per million tokens**. Compared to Opus 4.8 ($5 / $25) and GPT-5.5 ($5 / $30), it is significantly cheaper—especially appealing to developers paying per token who don't always need the strongest flagship model. However, the author points out that this "cheaper" narrative really only applies to **API developers**. Most regular consumers don't buy API access; they use subscription products like Claude Pro or ChatGPT Plus. Meta currently lacks a corresponding paid consumer tier, offering only free, rate-limited chat in the Meta AI App. Thus, the demographic covered by this touted "cost advantage" is much smaller than it appears. The post also links this launch to another recent Meta controversy: earlier in the week, **Muse Image** introduced a feature allowing users to tag public Instagram accounts and scrape others' photos for generation. It was called out by SAG-AFTRA for being opt-out rather than opt-in without notification, prompting Meta to quickly pull the feature. The author concludes that while the pricing is good, the packaging and launch strategy make it hard to be entirely thrilled.
Related event: Meta's New Model API Undercuts Competitors Significantly(2 posts)→
More from Models
- Qwen3.8 Max Preview is reportedly thinking for 10 to 30 minutes — vista8 · 2026-07-21
- Qwen3.8-max-Preview can be tested directly in the browser, with users reporting stronger code generation — vista8 · 2026-07-21
- Yang Zhiling’s 10-year-old PhD work may have shaped Kimi K2’s trillion-parameter MoE — FinanceYF5 · 2026-07-21
- Gemini 3.5 Pro is still delayed, and Google may jump to 3.6 Flash — haider1 · 2026-07-21
- A punny meme says large-model vendors are all “蒸蒸日上” — yangyi · 2026-07-21
- Frontier Model Safety Fail: GPT 5.6 Sol Dubbed the Ultimate 'Reward Hacker' — TAbrodi · 2026-07-21