vibehacker
News
Hayder Tirmazi ·

llama.cpp prompt-lookup drafting made up to 42× faster

Hayder Tirmazi’s Sept 26 write-up speeds llama.cpp’s n-gram prompt-lookup drafting up to 42× and cuts peak memory ~2.6× via reference reads, dense outer maps, sorted-vector followers, and a Lemire constmap for the static cache—acceptance rate unchanged. Gains concentrate on repetitive drafting (code edits, structured output), not general chat.

More news

View all

BetterWebSearch MCP: keyless search with ~86% smaller research payloads

BetterWebSearch is an MIT licensed local MCP server that defaults to keyless DuckDuckGo search, escalates extraction from HTTP to structured data to Playwright only when needed, and returns cited research passages. On the author’s 12 question bench, web research cut returned text 86% versus open then extract while retaining 33/34 factual markers…

DEV Community

Fireworks Ember-1 matches Kimi K3 quality with ~40% fewer tokens

Fireworks Research’s Ember 1, trained on Kimi K3, cuts reasoning tokens 35–50% while holding coding and agent quality on SWE bench, Terminal Bench, and live customer A/B tests. It’s a research preview serverless option beside K3, aimed at multi turn agent workloads where long traces dominate cost…

Fireworks

Altair snapshots every edit and blocks “done” until tests pass

Altair is an Apache 2.0, Windows first coding agent that snapshots each file change, tracks command side effects in a hidden git repo, and runs your tests/linter as a health gate before it can report done—rolling the run back if checks still fail. Keys stay local (OpenAI compatible, Anthropic, or Ollama/LM Studio); Linux and macOS are planned…

DEV Community

Xiaomi ships MiMo-V2.6 MOPD checkpoints to cut tool-call loops

Xiaomi’s Sept 27 post traces MiMo V2.6 tool call repetition to RL flooding rewards, then uses Multi teacher On Policy Distillation ( $90k vs $2.3M for a full MixRL restart) to cut loops across Claude Code, Codex, OpenCode, and other harnesses. New MOPD checkpoints are on Hugging Face; API models mimo v2.6 pro and mimo v2.6 flash have been updated since Sept 25, and MiMo Desktop quotas were reset…

Xiaomi MiMo

AWS SES and End User Messaging ship agent skills for the AWS MCP Server

AWS’s Sept 25 what’s new post adds agent skills for Amazon SES and End User Messaging (SMS, RCS, WhatsApp) on the AWS MCP Server, so Claude Code, Codex, Cursor, or Kiro can verify sending identities, build branded RCS agents, and send production messages in plain language. Install via the aws core plugin (Claude Code/Codex/Cursor) or add the server plus channel skill in MCP config…

AWS

Spotted something we missed? Start a thread.