MIT CSAIL's JAZ: minimalist agent with one 'invoke' primitive beats MemGPT at half the cost
omarsar0 · x · 2026-09-26
A new MIT CSAIL paper presents JAZ, a minimal agent framework with a single primitive: invoke. The LLM writes code, can call invoke recursively, and sees all its inputs and history as variables in the code environment.
Key results:
- With prompting only and no memory subsystem, it beats Letta (MemGPT) by 8% at half the cost on the recall-heavy part of StuLife
- On AppWorld it beats ACE, a self-improvement method, by 4% at lower cost
The core idea: memory and self-improvement are usually built as separate subsystems, but here both are code the agent writes inside the loop, with hooks enforcing constraints and monitoring. Strong empirical support for keeping agent harnesses minimal.
More from coding & agent
- Tiny 4B model drives a delivery car: one decision per junction, no 500-token monologue — MaziyarPanahi · 2026-09-26
- Stop Burning Credits: Why You Should Start Every Lovable Project in Chat Mode — damienghader · 2026-09-26
- skill-audit: Open-source pre-install auditor targets rising agent skill supply-chain risk — masiha97 · 2026-09-26
- SimonW: coding agents make software engineering harder, demanding extraordinary discipline — blaizedsouza · 2026-09-26
- Gemma 4 dev-agent comp locks everyone to one 31B model — is the code graph the intended edge? — politefella0 · 2026-09-26
- Power user: Opus 5.5 beat Fable 5.1 on cost and quality, one day of 5 subagents = 30min on Fable — breukr · 2026-09-26