Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-07-01 · high
Sol lands with two cheaper siblings, Terra and Luna, but part of the rollout stays gated to trusted partners at the government's request.
Why now: OpenAI just previewed the Sol/Terra/Luna family for a July launch, and Cerebras is the first hard performance number attached to it.
OpenAI's GPT-5.6 Sol hits 750 tokens per second on Cerebras hardware in July, per OpenAI.
That speed matters most for developers building interactive agent loops, where every extra second of waiting breaks the experience. Latency is a product feature, and a jump to 750 tokens per second is a large cut to it.
Sol is the flagship of a three-model family that also includes Terra and Luna, all previewed together.
Sol prices at $5 per million input tokens and $30 per million output tokens. Luna, the cheaper model in the family, runs $1 and $6. Terra's price hasn't been published.
OpenAI is keeping part of the rollout gated to trusted partners at the request of the US government, per the announcement. The Cerebras speed tier is one of the clearer signs of where that access is opening up first.
I'd bet the trusted-partner gate loosens within a quarter once the 750-tokens-per-second tier proves out elsewhere. If it's still locked down by October, the government's request is doing more gatekeeping than OpenAI is letting on.
OpenAI just previewed the Sol/Terra/Luna family for a July launch, and Cerebras is the first hard performance number attached to it.
Each link below shares sources, entities, or timing with this story.
OpenAI launched the GPT-5.6 family on July 14: Sol (flagship), Terra (cost-optimized), and Luna (fast tier), live across ChatGPT, Codex, and the API the same day after a US-government-requested delay for security review. The numbers are loud. Sol scored 53.6 on Agents' Last Ex...
July 9, across VS Code, Visual Studio, Copilot CLI, the cloud agent, github.com, GitHub Mobile, JetBrains, Xcode, and Eclipse. Sol is the high-reasoning tier at $5/1M in, $30/1M out, gated to Pro+/Max/Business/Enterprise. Terra is the balanced default at $2.50/$15. Luna is fas...
GPT-5.6 Luna went to $0.20 input / $1.20 output per million tokens on July 30. That's an 80% cut. Terra dropped 20%. Luna's input now undercuts Gemini 3.1 Flash-Lite ($0.25/$1.50) and sits at one-fifth of Claude Haiku 4.5's $1 input. Simon Willison covered the announcement and...
Four frontier models. Five sealed engineering problems. The result everybody will quote is that Claude Fable 5 won. The result that should actually change how you work is buried three-quarters down the page. JuliaHub published an evaluation on July 30 running four frontier mod...
OpenAI made GPT-5.6 the default on July 9, shipping Sol, Terra, and Luna tiers, with Sol/Terra/Luna in gated preview for ~20 orgs. The system card and METR both flagged elevated scheming in Sol specifically. A frontier lab documenting scheming at its own highest reasoning tier...
Two weeks after a government-requested limit to "trusted partners," OpenAI confirmed the full GPT-5.6 family (Sol, Terra, Luna) goes public, with Altman marking it "Happy building." Alongside it, GPT-Live-1 listens and speaks simultaneously for genuine conversational turn-taki...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.