DeepSeek V4.1-Flash cuts KV cache memory to a quarter, targets cheaper agents
The Decoder · rss · 2026-09-10
- DeepSeek released V4.1-Flash, a multimodal model with 552B total parameters and only 16B active per token.
- The key optimization cuts KV cache memory to 1/4 of its predecessor, aimed at making AI agents far cheaper to run.
- It narrowly beats Opus 5 and GPT-5.6 Sol on the DeepSWE coding benchmark.
- The model ships under the MIT license.
More from Infra
- Engineer with thousands of fabricated devices pours cold water on photonic crystal modulator yields — jwt0625 · 2026-10-26
- Anthropic S-1 reveals 48 cents of compute cost per dollar of Q2 revenue — nmasc_ · 2026-10-26
- Persistent AI joins UR network to route private AI inference with encrypted, IP-hidden traffic — 0xSammy · 2026-10-26
- Anthropic's disclosed compute commitments: $114.5B Google, $161.2B Broadcom top the list — BenBajarin · 2026-10-26
- SpaceX strikes $40B deal with Apollo Global to lease AI chips and data center hardware — Polymarket · 2026-10-26
- SpaceX AI revenue up 830%, Claude designs 318K antivenom proteins in 64 hours — PeterDiamandis · 2026-10-26