cognee vs hindsight

The two are close in size: 31.0k stars for cognee, 30.5k for hindsight. Over the days we have tracked them cognee moved +21.6% and hindsight +72.7%, so hindsight is growing faster right now.

They split the axes: cognee leads on context depth, hindsight on setup ease.

Stars and commit dates come from our own daily tracking. The six axes are read off each project's documentation by our review pipeline, so they describe what a project says about itself, not what we measured in its code.

Where they stand today

Runs the full memory layer (graph, vectors, sessions) on a single self-hosted Postgres-backed stack, combining vector search and evolving knowledge graphs.

Stars
31.0k
Tracked growth
+21.6%
Maturity
●●●●●
Last commit
9h ago
Language
Python
License
Apache-2.0
Cost to run
Your LLM API key, or use Cognee Cloud (managed)

Hindsight provides a learning-focused agent memory with biomimetic data structures and a reflect operation to build mental models, going beyond simple RAG-style recall.

Stars
30.5k
Tracked growth
+72.7%
Maturity
●●●●●
Last commit
14h ago
Language
Python
License
MIT
Cost to run
Your LLM API key (or self-hosted models); Hindsight itself can be self-hosted
0%+73%90 tracked days
topoteretes/cogneevectorize-io/hindsight

Six axes, head to head

Each axis runs 0 to 5. The label under a score is what that project's own docs claim, not a category average.

Axiscogneehindsight
Context depth
How much of your codebase it sees before it answers: the open diff, the diff plus related files, or the whole repository.
●●●●●
Whole-graph memory
●●●●●
Related context
Noise control
How it keeps output volume down — severity thresholds, deduplication, incremental runs over new commits only.
●●●●●
Feedback & routing
●●●●●
Rerank & filtering
Customization
How far it bends to your team: custom rules, prompts, style guides, per-path config.
●●●●●
Config + plugins
●●●●●
Config API & metadata
Privacy
Whether your code stays on your own infrastructure: fully local, self-hostable, or cloud API only.
●●●●●
Self-hostable
●●●●●
Self-hostable
Model freedom
Whether you can point it at any provider, or it is wired to one.
●●●●●
Bring-your-key
●●●●●
Bring-your-own / local
Setup ease
What it takes to get a first useful run out of it.
●●●●●
API key required
●●●●●
One-command start

Which one to pick

Pick cognee if…

Self-hostable — run a unified, persistent agent memory on a single Postgres instance with minimal extra services and strong integration options.

  • Context depth: Whole-graph memory (5/5 against 3/5)
Runs in cli, web-app, coding-agent-plugin. Works with byok, openai.

Pick hindsight if…

Learning-focused — choose Hindsight when you need agents that form and reflect on long-term memories (mental models) with easy self-hosted deployment and multi-provider model support.

  • Setup ease: One-command start (5/5 against 3/5)
Runs in cli, web-app, coding-agent-plugin. Works with byok, openai, anthropic, gemini, local-ollama, other-fixed.

What people want from each one

Questions people ask

Is cognee better than hindsight?

They split the axes: cognee leads on context depth, hindsight on setup ease. cognee is worth picking when self-hostable — run a unified, persistent agent memory on a single Postgres instance with minimal extra services and strong integration options.

Which of cognee and hindsight keeps my code private?

cognee: Self-hostable (4/5). hindsight: Self-hostable (4/5).

What does each one cost to run?

cognee: Your LLM API key, or use Cognee Cloud (managed). hindsight: Your LLM API key (or self-hosted models); Hindsight itself can be self-hosted.

Full profiles: topoteretes/cognee and vectorize-io/hindsight. Everything else in Memory & context.