AI model daily brief: October 20, 2026 — Hy3 gets cheaper

MiniMax and DeepSeek are experiencing major service disruptions, while Mistral AI has doubled its pricing for Mistral Large 4. Developers should prioritize reliability checks and review cost adjustments for affected models before deploying to production.

Outages & incidents

• Fireworks AI — Service Degradation for GLM 5.3 (impact: minor) · status
• Fireworks AI — Service Degradation for Deepseek V4.1 Flash (impact: minor) · status
• Fireworks AI — Service Degradation for Kimi K3 (impact: minor) · status
• DeepSeek — 搜索服务异常（Search service error） (impact: major) · status
• MiniMax — Elevated errors on Large Language Models (LLM) (impact: major) · status
• MiniMax — Elevated errors on Large Language Models (LLM) (impact: major) · status
• MiniMax — Elevated errors on Large Language Models (LLM) (impact: major) · status
• MiniMax — Elevated errors on Large Language Models (LLM) (impact: major) · status
• MiniMax — Elevated errors on Large Language Models (LLM) (impact: major) · status
• Fireworks AI — Elevated latency on some serverless models · status
• MiniMax — Elevated errors on Large Language Models (LLM) (impact: major) · status

Reliability watch

• MiniMax: 45 failed and 42 degraded checks out of 711
• DeepSeek: 5 failed and 0 degraded checks out of 712
• Kimi (Moonshot AI): 0 failed and 45 degraded checks out of 711
• Fireworks AI: 0 failed and 31 degraded checks out of 711
• Azure OpenAI / Microsoft Foundry: 0 failed and 2 degraded checks out of 711

Price changes

• Mistral Large 4 (Mistral AI): $0.68 / $2.09 → $1.36 / $4.18 (+100%)
• Qwen3 Next 80B A3B Instruct (Qwen (Alibaba)): $0.09 / $1.10 → $0.10 / $1.10 (0%)
• Hy3 (Tencent): $0.13 / $0.53 → $0.0825 / $0.33 (-37%)
• Hy4 preview (Tencent): $0.83 / $2.50 → $0.75 / $2.25 (-10%)
• DeepSeek V4.1 Flash (DeepSeek): $0.0355 / $0.12 → $0.0283 / $0.12 (0%)
• DeepSeek V4 Flash 0731 (DeepSeek): $0.44 / $1.32 → $0.13 / $0.39 (-70%)
• DeepSeek V4 Pro 0423 (DeepSeek): $0.96 / $1.91 → $0.38 / $0.76 (-60%)
• GLM 5 (Z.ai (Zhipu)): $0.60 / $1.92 → $1.00 / $3.20 (+67%)
• DeepSeek V4.1 Flash (DeepSeek): $0.0352 / $0.0112 → $0.0355 / $0.12 (+971%)
• Gemma 4 26B A4B (Google): $0.0765 / $0.26 → $0.09 / $0.30 (+18%)
• Qwen3.8 27B (Qwen (Alibaba)): $0.24 / $2.20 → $0.04 / $1.35 (-39%)
• Qwen3.6 35B A3B (Qwen (Alibaba)): $0.15 / $1.00 → $0.10 / $0.95 (-5%)
• DeepSeek V3 (DeepSeek): $0.32 / $0.89 → $0.26 / $1.03 (+16%)
• Gemma 3 4B (Google): $0.05 / $0.10 → $0.04 / $0.05 (-50%)
• GLM 5 (Z.ai (Zhipu)): $1.00 / $3.20 → $0.60 / $1.92 (-40%)

Deprecations & removals

• Llama 3.3 Euryale 70B was removed from the catalog
• R1 was removed from the catalog
• WizardLM-2 8x22B was removed from the catalog

This brief is generated automatically from ModelBank's live monitoring data. Want these changes the moment they happen? Get free alerts.