Plan event · 17 Sept 2026
All plan events · 31 providers tracked
CommandCodeceiling
What moved
Calculator rules modeled every GOAT model at the plan’s full $70 credit pool and every Pro model at $80 — CommandCode’s published per-model credit allowances were never applied → Per-model monthly credit allowances applied (a model may draw at most its own allowance from the pool): GOAT $20 standard, $30–$33 on the MiMo / Qwen3.6-Plus / Qwen3.7 rows, $40 on the Flash rows (Gemini 3.7/3.8 Flash, GLM-5.3 Flash), $47 MiniMax M3, $60 on the DeepSeek/Kimi-K2.7-Code rows, up to the full $70 on the negotiated rows (GPT-5.6 Sol, GLM-5.2, Tencent Hy3, Qwen 3.8 27B); Pro the same shape at $30 / $40–$43 / $50 / $57 / $70 / $80
Site correction: CommandCode GOAT and Pro credits are per-model allowances, not one shared pool, so the calculator was overstating most rows — a model with a $20 allowance was modeled as if it could spend GOAT’s whole $70 (Pro’s $80). Every GOAT and Pro row now budgets min(plan pool, published allowance), which reproduces the vendor’s own request estimates (DeepSeek V4.1 Flash: 154,000 req/mo on GOAT at its $60 allowance). DeepSeek V4.1 Flash’s allowance is $40 GOAT / $50 Pro, boosted to $60/$70 through 2026-09-20. Rates, plan prices and credit grants are unchanged.
Verified 2026-09-17 · source
Affected plans
More CommandCode events
- 2026-09-30 · Models — CommandCode now serves a second fast-tier DeepSeek row, DeepSeek V4.1 Flash Fast, on every
- 2026-09-30 · Models — CommandCode shipped GPT-6.1 Sol the day OpenAI launched it (changelog 1.71.0, Add GPT-6.1
- 2026-09-30 · Models — CommandCode now serves Claude Sonnet 5.5 from GOAT upward. The provider pricing-limits pag
- 2026-09-28 · Models — CommandCode serves four paid models on every plan, including the $1 Go tier, that no track