Plan event · 28 Oct 2026
All plan events · 31 providers tracked
Standard Computemodels
What moved
34-model smart-routed catalogue (14 listed) → 46-model smart-routed catalogue, incl. GPT-6.1 Sol / Terra, Gemini 4 Argon, Claude Sonnet 5.5 / Haiku 5.5, Kimi K3.1 Pro, MiniMax M3.1 Flash, Grok 4.3, Mistral Large 4 and the GPT-5.6 family
The /models page now serves 46 distinct models (32 further EU/US rows are key-scope variants of the same models) instead of 34, and every plan publishes the whole catalogue. The additions are the GPT-6.1 and GPT-5.6 families, GPT-6 Luna, Gemini 4 Argon and the Gemini 2.5 pair, the Qwen 3.7 line plus Qwen 3.8 Flash, MiMo V2.5, Kimi K3.1 Pro, MiniMax M3.1 Flash, Grok 4.3, Mistral Large 4 and the rest of the Claude 5.5 family. Prices and compute budgets are unchanged.
Verified 2026-10-28 · source
Affected plans
More Standard Compute events
- 2026-10-28 · Ceiling — The first-month launch offer is gone: the pricing page no longer carries "1.5", "1.5x", "l
- 2026-09-24 · Price — Cache-provenance re-verify of the Standard Compute rule: the provider resells at list rate
- 2026-09-12 · Launch — Standard Compute added to the catalogue: an OpenAI-compatible, smart-routed flat-rate API