The week's launches, dated:
| Model | Vendor | Date | Price (in / out, per M tokens) | Note |
|---|---|---|---|---|
| GPT-5.6 | OpenAI | Jul 9 | tiered (Sol/Terra/Luna) | Now the ChatGPT default |
| Grok 4.5 | xAI | Jul 8 | $2 / $6 | 500K context window |
| Muse Spark 1.1 | Meta | Jul 9 | $1.25 / $4.25 | Proprietary metered API |
| Seedream 5.0 Pro | ByteDance | Jul (wk) | — | Image generation |
OpenAI moved GPT-5.6 to general availability on July 9 and made it the ChatGPT default, splitting the family into Sol, Terra, and Luna (TechCrunch). A day earlier, xAI shipped Grok 4.5 at $2 per million input tokens, $6 output, with a 500K context window (pricepertoken). ByteDance put out Seedream 5.0 Pro (llm-stats).
Meta's launch differs from its open-weight Llama strategy: Muse Spark 1.1 is proprietary and metered through the Meta Model API.
Muse Spark 1.1 lists at $1.25 per million input tokens and $4.25 per million output tokens (Artificial Analysis). The proprietary Muse line differs from Meta's open-weight Llama releases.
At listed rates, Muse Spark 1.1 undercuts Grok 4.5 on both input and output tokens. OpenAI's GPT-5.6 pricing varies by model tier.
The listed price makes Muse Spark 1.1 a relatively inexpensive hosted model. Capability testing, utilization, and operating costs still determine whether API use beats self-hosting for a particular workload.
Watch whether later Muse releases remain API-only or gain downloadable weights.
The Signal is the public edge of a private practice. Sherpa points the same intelligence engine at one owner's business — competitors, suppliers, regulators, watched daily, graded and sourced. Work with a Sherpa →