Plan event · 25 Sept 2026
All plan events · 31 providers tracked
Ollamamodels
What moved
Pro and Max listed 20 Ollama Cloud models, including deepseek-v4-flash, glm-5.1 and qwen3.5:397b — all three with published per-token rows → Pro and Max list 17 cloud models — deepseek-v4-flash and glm-5.1 retired on 2026-09-25, qwen3.5:397b no longer cloud-enabled; every remaining rate unchanged
Three models leave the Ollama Cloud roster between the 2026-09-24 and 2026-09-26 pricing dumps: the deepseek-v4-flash, glm-5.1 and qwen3.5:397b rows are gone from ollama.com/pricing and no other row changed. Ollama’s own library pages state deepseek-v4-flash and glm-5.1 “was retired on September 25, 2026”; qwen3.5 keeps a library page but has lost its cloud tag — it is absent from the cloud lineup and from the public /api/tags cloud list. Both tracked plans drop the three from their model lists and the calculator loses their per-token entries, so GLM-5.1’s mid-priced GLM tier is no longer reachable on Ollama (GLM-5.2 at $1.40 / $4.40 and GLM-5.3-Flash at $0.15 / $0.50 remain).
Verified 2026-09-26 · source
Affected plans
More Ollama events
- 2026-09-24 · Models — Ollama Cloud adds deepseek-v4.1-flash to the model roster and to both tracked plans (Pro a
- 2026-09-24 · Price — Ollama withdraws the cached-input rates for mistral-large-3, nemotron-3-nano and qwen3.5:3
- 2026-09-02 · Ceiling — Ollama Cloud moved from multipliers to a credits model and unpaused Max. Pro ($20) and Max