DeepSeek V4 Flash drops with 284B parameters

// 45d agoMODEL RELEASE

DeepSeek V4 Flash drops with 284B parameters

DeepSeek releases V4 Flash, a 284B parameter MoE model that achieves high-speed reasoning with only 13B active parameters. Its novel architecture uses native FP4/FP8 mixed precision and "Hybrid Attention" to drastically reduce VRAM and KV cache requirements.

// ANALYSIS

DeepSeek is redefining efficiency by proving that massive parameter counts don't require massive hardware if you're smart about compression and attention.

–Native FP4/FP8 storage means the "unquantized" weights are already as small as typical 4-bit quants of other models
–Hybrid Attention architecture reduces the KV cache memory footprint by 90% compared to previous generations
–With only 13B active parameters, it offers GPT-4 class reasoning speeds on consumer-grade (multi-GPU) or Mac Studio hardware
–The release of weights under MIT license continues DeepSeek's trend of disrupting the proprietary model landscape
–"Think" modes allow developers to trade off latency for deeper reasoning depth depending on the task

// TAGS

deepseek-v4-flashllmmoeopen-weightsreasoninginferenceinfrastructure

DISCOVERED

45d ago

2026-04-27

PUBLISHED

45d ago

2026-04-26

RELEVANCE

10/ 10

AUTHOR

WyattTheSkid

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

LAUNCH27m ago

New framework automates agent harness optimization

A developer has built a framework designed to automate the optimization of AI agent harnesses—the control scaffolding that manages planning, tool calls, work verification, and error recovery around large language models. The framework's optimization capabilities were demonstrated on Terminal-Bench, a highly demanding benchmark for terminal-based agentic tasks.

OPEN SOURCE43m ago

Activeloop has open-sourced Hivemind, a continual learning skill layer for AI coding agents that watches task traces to automatically generate and share reusable skills across team environments.

Hivemind is an open-source tool developed by Activeloop that enables AI coding agents (such as Claude Code, Cursor, and Codex) to learn from user actions and repeat patterns instead of relying solely on retrieving static facts. By monitoring agent traces, including prompts, tool calls, and responses, Hivemind automatically identifies recurring patterns and distills them into reusable skill files (e.g., as slash commands). These skills can then propagate across a development team's shared trace store, facilitating collaborative agent improvement. Using Deep Lake for structured trace storage, Hivemind supports custom cloud configurations (BYOC) and exhibits substantial performance improvements, reducing token usage and turn counts on long-context benchmarks.

MODEL46m ago

Claude Fable 5 launches on B.AI

Anthropic's newly released Claude Fable 5 model is now available on the B.AI platform, offering unified access and smart routing for the long-running agentic model. Optimized for multi-hour, highly uncertain workflows, the model is designed to power autonomous agents executing software engineering and scientific research tasks.