Galahad is long-term memory for AI agents, distributed as the galahad-kv Python package on PyPI. It is built around a very large context window for agents — the launch pitch is a 50 million token window — and it works with vLLM and SGLang inference servers. The underlying research is published on arXiv, covering byte-exact deduplication in retrieval-augmented generation, a deterministic deduplication method called Merlin, and byte-exact KV-cache grafting described as a verified-knowledge flywheel.
Be the first to rate Galahad. Your feedback helps other users decide.
Add galahad-kv from PyPI to a project that serves models through vLLM or SGLang.
Give an agent memory that reaches far past a single context window.
Follow the arXiv papers behind the deduplication and KV-cache grafting approach.
Similar tools in the same category — compare side-by-side
Hapsland catches questionable data and code decisions while your coding agent is still working. It a...
SpecWeave is intent-first development: settle what you want before the code, hand the work between C...
Singularity is memory for coding agents. When an agent commits a change it records the task, the fil...
Guides and comparisons about code tools
Ten launches that keep the model, the index and the facts on your own machine, from search that admits no match to GPU training in CUDA.
Cursor vs Zencoder vs Greptile compared for 2026: autocomplete, repo-level agents and AI code review. Which AI coding tool fits how your team actually works?
A practical comparison of the coding agents developers actually use in 2026: asynchronous cloud, terminal and IDE agents, and how to choose.
Compare Cursor, GitHub Copilot and Devin to find the best AI coding assistant. Learn how each fits different developer workflows, plus Cody and Codeium.
Build mobile apps faster with 7 AI code tools in 2026. From AI-assisted coding and UI generation to automated testing and deployment.
