A tiny C program where an LLM rewires its DAG while running
1000-line C program where an LLM can edit the DAG mid-run, but only via a four-verb grammar, and every edit is an auditable event - kouhxp/liteflow
A tiny C program where an LLM rewires its DAG while running launched on May 5, 2026 with 15 votes, #1,002 of 20,538 launches that month and more than 90% of that year's launches.
In plain words
This is a C program that allows a large language model to modify its computational graph during execution. The program is designed for developers and researchers exploring dynamic neural network architectures. What stands out is the ability for an LLM to rewire its own directed acyclic graph in real-time, potentially enabling more adaptive and flexible model behavior. The implementation is kept deliberately minimal, written in C for efficiency and accessibility.
written from the facts on this page · September 2026
From the sources
A ~1000-line C program that runs YAML-defined DAGs, where an LLM can edit the graph mid-run. When a task fails, a planner LLM gets the stderr and emits one of four verbs: RETRY , PATCH , INSERT_BEFORE , ABORT . The runtime applies the mutation and continues. Every edit is recorded in an append-only event log, so you can replay any run and see exactly which graph changes the LLM made and why. Not an agent. A runtime where the LLM is a peer of the scheduler. cc -std=c11 -O2 -o liteflow liteflow.c Single file, no libraries. Shells out to curl for HTTP. Linux, macOS, modern Windows. export OPENAI_API_KEY=... export OPENAI_BASE_URL=https://api.deepinfra.com/v1/openai # any OpenAI-compatible…from github.com
Does a similar job
all alternatives →
I built a tiny LLM to demystify how language models workApr 2026 · github.com · ▲915Built a ~9M param LLM from scratch to understand how they actually work. Vanilla transformer, 60K synthetic conversations, ~130 lines of PyTorch. Trains in 5 min on a free Colab T4. The fish thinks the meaning of life is food. Fork it and swap the personality for your own character.
Tiny-vLLM – high performance LLM inference engine in C++ and CUDAMay 2026 · github.com · ▲205Build your own high performance LLM inference engine in C++ and CUDA - a smaller version of vLLM - jmaczan/tiny-vllm
Kent Dybvig's Scheme Machine in 400 Lines of C (Heap-Memory Model)2025 · gist.github.com · ▲226Heap based scheme machine. GitHub Gist: instantly share code, notes, and snippets.
Pipelex – Declarative language for repeatable AI workflowsOct 2025 · github.com · ▲122We’re Robin, Louis, and Thomas. Pipelex is a DSL and a Python runtime for repeatable AI workflows. Think Dockerfile/SQL for multi-step LLM pipelines: you declare steps and interfaces; any model/provider can fill them. Why this instead of yet another workflow builder? - Declarative, not glue code: you state what to do; the runtime figures out how. - Agent-first: each step carries natural-language context (purpose, inputs/outputs with meaning) so LLMs can follow, audit, and optimize. Our MCP server enables agents to run pipelines but also to build new pipelines on demand. - Open…
Fine-grained stylistic control of LLMs using model arithmetic2023 · github.com · ▲85We developed a new framework that enables flexible control of generated text in language models. By combining several models and/or system prompts in one mathematical formula, it lets you tweak your style and combine model outputs with ease. A handy tool for those working with LLMs, looking for more fine-grained control of stylistic output. More details in our paper: https://arxiv.org/abs/2311.14479. Feedback and potential applications are welcome.
LLM Verified with Monte Carlo Tree Search2023 · github.com · ▲102This is a weekend hack that I'd like to further develop as it's working surprisingly well. Using MCTS, we can explore a space of possible verified programs with an LLM. We check the partial programs at each step, and so steer towards programs that pass the verifier. https://github.com/namin/llm-verified-with-monte-carlo-tree-...
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 21d ago · stillwet.art
PixVerse R2▲381A real-time world model you can explore and change
AI · 30d ago · world.pixverse.video
Source: Show HN post ↗
Launched alongside, May 2026
the whole month →
Parallel agents, diff reviewer, and multi-model comparisons
Dev tools · May 2026 · kilo.ai
Hey HN, Henry here from Cactus. We open-sourced Needle, a 26M parameter function-calling (tool use) model. It runs at 6000 tok/s prefill and 1200 tok/s decode on consumer devices. We were always frustrated by the little effort made towards building agentic models that run on budget phones, so we conducted investigations that led to an observation: agentic experiences are built upon tool calling, and massive models are overkill for it. Tool calling is fundamentally retrieval-and-assembly (match query to tool name, extract argument values, emit JSON), not reasoning. Cross-attention…
Life & fun · May 2026 · github.com
🌱 Private, quiet space for thinking. Simple app for .md files. - zakirullin/files.md
Dev tools · May 2026 · github.com