DeepSeek-V4.1-Flash Hits the API: 1/8 the SSD Cache, V4-Pro Being Phased Out
deepseek_ai · x · 2026-09-10
DeepSeek announced V4.1-Flash is now live on the DeepSeek API (model name deepseek-flash) with native multimodal support, plus migration and cost-cutting details:
- Old models retired: V4-Flash and V4-Flash-Vision-Exp are retired; legacy model names temporarily route to V4.1-Flash
- Independent tests put V4.1-Flash ahead of flagship V4-Pro on performance, cost, speed, and total runtime — V4-Pro is being phased out. From Sept 14, 04:00 UTC 2026, all deepseek-v4-pro requests route to V4.1-Flash at V4.1-Flash rates until V4.1-Pro launches
- Massive KV cache compression: just 1/4 the HBM and 1/8 the SSD storage vs the previous generation. DeepSeek notes cache-hit charges are a large share of agent costs, so compression cuts those significantly
- Official partners WorkBuddyAI (including Codebuddy) and opencode fully support V4.1-Flash
More from Infra
- DeepSpeed shows AutoEP/AutoTP/AutoSP parallelism gains, hosts office hours — PyTorch · 2026-10-27
- Cloudflare Durable Objects validated as top architecture for AI agent WebSocket servers — tristanbob · 2026-10-27
- Gary Marcus: With Q3 data added, Anthropic's compute-efficiency growth shows deceleration — GaryMarcus · 2026-10-27
- Inference is just Kubernetes with a multi-trillion-dollar market, and the communities overlap — matt_slotnick · 2026-10-27
- vLLM eyes Rust migration as inference engines race to rewrite the stack — AccBalanced · 2026-10-27
- Resin lets an agent build standalone LLM inference engines; vLLM and SGLang optimize DeepSeek kernels differently — AccBalanced · 2026-10-27