It does three things:
- It works as a drop-in cache. It speaks the same protocol as Redis, so your existing Redis client and code keep working with no changes.
- It can store and search vectors, and it can cache AI answers by meaning. So if two prompts mean the same thing, RAMen can return the saved answer instead of calling the model again. That saves you money.
- It has a built-in MCP server. That means AI agents (like Claude) can read, write, search, and remember data in RAMen directly as a tool. No extra glue code needed.
Redis was built for app servers. RAMen is built for AI agents.
Is this a Redis alternative?
Yes, for a lot of common use cases. If you use Redis (or Valkey) as a cache or a simple key value store, you can point your app at RAMen instead and it will just work, because RAMen speaks the Redis protocol (RESP2).
Where RAMen is different:
- It has vector search and a semantic cache built in. With Redis you would need an extra module or extra code for that.
- It has an MCP server built in, so AI agents can use it as a tool out of the box.
- It is one small single file (a single binary) with no extra dependencies. Easy to download and run.
What RAMen does NOT do yet (so you know what you are getting):
- No clustering, no replication, no failover.
- No append-only-file durability (it saves snapshots to disk instead).
- Not tuned to beat Redis or Valkey on raw speed.
So: great as a Redis-style cache and an AI data layer for one machine. Not yet a replacement for a big production Redis cluster.
Quick start (the simple version)
You need Go installed (version 1.25 or newer). Then run one of these.
Option 1: install with Go
go install github.com/Rohit-Dnath/RAMen/cmd/ramen@latest
ramen
Option 2: run with Docker
docker run -p 6379:6379 -p 8080:8080 ghcr.io/rohit-dnath/ramen:latest
Option 3: build from source
git clone https://github.com/Rohit-Dnath/RAMen
cd RAMen
go run ./cmd/ramen
When it starts you will see:
ramen: listening on [::]:6379
ramen: dashboard on http://localhost:8080
That is it. RAMen is now running on port 6379, and a live dashboard is at http://localhost:8080.
How to use it
Talk to it like Redis
If you have redis-cli installed, just connect to port 6379:
redis-cli -p 6379
> SET hello world
OK
> GET hello
"world"
> EXPIRE hello 60
(integer) 1
If you have an app that already uses Redis, point it at RAMen. Usually you only change one setting:
export REDIS_URL=redis://localhost:6379
Basic commands
These work just like Redis:
SET user:1 "Rohit" store a value
GET user:1 read it back
INCR visits count something
HSET person name Rohit store fields under one key
RPUSH queue job1 job2 a list
SADD tags go ai a set
ZADD board 100 alice a ranked list
SUBSCRIBE news listen for messages
PUBLISH news "hello" send a message
Full list with examples: docs/commands.md.
Vector search
Store vectors and find the closest ones. Your app gives RAMen the numbers, no API key needed.
VSET docs d1 0.1 0.2 0.9 META "intro page"
VSET docs d2 0.9 0.1 0.0 META "pricing page"
VSEARCH docs 0.1 0.2 0.85 TOPK 1 WITHSCORES
Semantic cache (save money on AI calls)
Cache an AI answer once. Next time a similar question comes in, get the saved answer back instead of paying for another model call.
SCACHE.SET "What is the capital of France?" "Paris"
SCACHE.GET "whats the capital of france" THRESHOLD 0.9
> "Paris"
This needs an embeddings provider. It works with a free local Ollama or with OpenAI. See Configuration.
Agent memory
Let an AI agent remember things across turns:
REMEMBER session7 user_name Rohit
RECALL session7 user_name
> "Rohit"
Use it from Claude (MCP)
Start RAMen, then add this to your Claude config and Claude can use RAMen as a tool:
{
"mcpServers": {
"ramen": {
"command": "ramen",
"args": ["mcp", "--addr", "localhost:6379"]
}
}
}
Step by step guide: docs/mcp.md.
The dashboard
Open http://localhost:8080 in your browser while RAMen is running. You will see your keys, memory use, and the cache hit rate, updating live. Turn it off with --dashboard-addr "".
Prometheus metrics
RAMen exposes its counters in Prometheus text format at http://localhost:8080/metrics, ready to scrape into Prometheus/Grafana with no extra setup (and no third-party dependency — it is hand-written). Everything is prefixed ramen_: ramen_commands_processed_total, ramen_connections_total, ramen_cache_hits_total / ramen_cache_misses_total / ramen_cache_hit_ratio, ramen_keys, ramen_memory_alloc_bytes, ramen_connected_clients, and ramen_uptime_seconds. The endpoint rides on the dashboard, so it is only served when the dashboard is enabled — with --dashboard-addr "" there is no /metrics to scrape. It reports counts only (no keys or values) and, like the rest of the dashboard, is unauthenticated.
RAMen vs Redis vs Valkey
Benchmarking
Configuration
Every flag also has a RAMEN_* environment variable.
Turn on the semantic cache for free with a local Ollama:
ollama pull nomic-embed-text
export RAMEN_EMBED_URL=http://localhost:11434/v1/embeddings
export RAMEN_EMBED_MODEL=nomic-embed-text
ramen
Or use OpenAI:
export RAMEN_EMBED_URL=https://api.openai.com/v1/embeddings
export RAMEN_EMBED_MODEL=text-embedding-3-small
export RAMEN_EMBED_KEY=sk-your-key
ramen
Roadmap
- Done: core Redis commands, strings, hashes, lists, sets, sorted sets, expiry, pub/sub, snapshot saving.
- Done: vector store, semantic cache, agent memory, MCP server, dashboard.
- Next: replication, stronger durability, clustering, faster vector index (HNSW).
Support the project
RAMen is free and open source, and it always will be. If it saves you time or you would like to help fund its development, you can support the work directly. It genuinely helps keep the project moving. Not in a position to chip in? A star, a share, or a bug report helps just as much. Thank you.
Contributing
RAMen is open source and very new, which means this is a great time to get involved and make a real difference. Beginners are welcome. You do not need to be a Go expert.
Good first things to help with:
- Try it out and report bugs or confusing parts.
- Improve the docs or add examples.
- Add a missing Redis command.
- Test RAMen with your favourite Redis client and tell us what worked.
Good first issues
New here? These are small, self-contained, and well described. Pick one, comment to claim it, and you are off.
- Good first issues beginner-friendly, clearly scoped tasks.
- Help wanted things we would love a hand with.
- All open issues the full list.
Most open issues are missing Redis commands with the exact files and acceptance criteria already written down, so you can follow an existing handler as a template and open a PR the same day.
Then read the Contributing Guide to get started. If you are not sure where to begin, open an issue and say hi. Stars and shares also help a lot.
Contributors
Thanks to these wonderful people who help make RAMen better.
</a>
</td>
<td align="center">
<a href="https://github.com/bayrambartu">
<img src="https://avatars.githubusercontent.com/u/113308882?v=4" width="80" height="80" style="border-radius:50%" alt="bayrambartu"><br>
<sub><b>bayrambartu</b></sub><br>
</a>
</td>
<td align="center">
<a href="https://github.com/tchalikanti1705">
<img src="https://avatars.githubusercontent.com/u/198486914?v=4" width="80" height="80" style="border-radius:50%" alt="tchalikanti1705"><br>
<sub><b>tchalikanti1705</b></sub><br>
</a>
</td>
<td align="center">
<a href="https://github.com/gaoharimran29-glitch">
<img src="https://avatars.githubusercontent.com/u/225884102?v=4" width="80" height="80" style="border-radius:50%" alt="gaoharimran29-glitch"><br>
<sub><b>gaoharimran29-glitch</b></sub><br>
</a>
</td>
Star History
License
BSD-3-Clause. This is the same license family Redis used before 2024, and the one Valkey uses. Simple and permissive, no surprises.