# Memory layer for AI agents that saves tokens, improves itself, reads decks, docs & sheets, enforces access control, ends RAG guesswork, runs on-prem, and never locks you in.

Source: <https://goodmem.ai/>

Title: Goodmem - Agentic AI Memory Infrastructure

> GoodMem is the memory layer for agentic AI — persistent, governed context for your agents, with retrieval quality validated on your own workload.

GoodMem connects your agents to your documents, your data, and their own history — and keeps making them better, automatically. Get answers that rival GPT and Claude at a fraction of the cost, and your data never has to leave your premises.

[Start free](https://cloud.goodmem.ai/login?loc=hero) [Self-host](https://goodmem.ai/quick-start)

[Compare deployment options ↓](https://goodmem.ai/#deployment)[Contact sales](https://goodmem.ai/contact-sales?interest=deployment)

Available in

- [ChatGPT](https://chatgpt.com/plugins/plugin_asdk_app_6aaae3ab7b1081918c3bb0921689fe15)
- [Claude](https://claude.com/plugins/goodmem)
- [Google ADK](https://google.github.io/adk-docs/integrations/goodmem/)

* [−53% Fewer tokens per query Retrieves the right context on demand, so agents finish in fewer steps. Median · production agent benchmark](https://goodmem.ai/resources/cutting-agent-token-burn)
* [+35% Sharper retrieval Automated fine-tuning adapts your retrieval models to your data. Median gain · 8 public retrieval tasks](https://goodmem.ai/resources/agent-optimization-architecture#retrieval-models)
* [35× Cheaper to run 91% of Claude Opus quality from a small model you run yourself. Per 1K conversations · BI-agent benchmark](https://goodmem.ai/resources/agent-optimization-architecture#llm)

Enterprise model training

## Frontier-class AI you can own

Price-performance comparison for a custom BI agent over 20 Oracle EBS scenarios. Only the model changes.

**Pass rate and cost per 1,000 conversations for four models on the BI-agent benchmark**

| Model                                 | Pass rate | Cost / 1K conv. |
| ------------------------------------- | --------- | --------------- |
| Claude Opus 4.8 API only              | 61.7%     | $2,380          |
| PAIR-trained 35B model 2×H100 you own | 55.9%     | $67             |
| GLM-5.2 16×H100                       | 54.4%     | $260            |
| GPT-OSS-120B 1×H100                   | 30.0%     | $10             |

Frontier APIs are smart but expensive, and every call leaves your network. Small models can be self‑hosted but don’t perform well without tuning. For one customer’s BI agent, PAIR trained a 35B model that reached 91% of Opus 4.8’s score at 1/35 of the cost, running on two H100s.

[Read the benchmark](https://goodmem.ai/resources/agent-optimization-architecture#llm) [Talk to us about a model for your agent](https://goodmem.ai/contact-sales?interest=benchmark)

Retrieval Optimizer

## Automatically selects the best retrieval stack for your data

The Retrieval Optimizer in GoodMem Cloud benchmarks embedder and reranker combinations — open-weight or API — on your real queries, then tells you which ones are statistically near-best. Evidence, not vibes.

- **Your queries, not a leaderboard** Every candidate is scored on your own workload and held-out data.

- **Honest uncertainty** 95% ranges and near-best odds, so you can trade a sliver of quality for cost or latency.

- **Keeps going** As your data and traffic change, GoodMem re-tunes the stack — no labeling required.

cloud.goodmem.ai · Retrieval Optimizer

Result: Pipelines 1 and 2 are strong choices.

1. Embedder: OpenAI Embedding 3 Large Reranker: Voyage Rerank 2.5 Mean NDCG 0.794 (95% range 0.760–0.825); 100% chance of being near-best, 100% chance of being best.
2. Embedder: OpenAI Embedding 3 Small Reranker: Voyage Rerank 2.5 Mean NDCG 0.787 (95% range 0.753–0.819); 100% chance of being near-best, 1% chance of being best.
3. Embedder: OpenAI Embedding 3 Large Reranker: Jina Reranker V3 Mean NDCG 0.739 (95% range 0.699–0.775); 31% chance of being near-best, 0% chance of being best.
4. Embedder: OpenAI Embedding 3 Small Reranker: None Mean NDCG 0.690 (95% range 0.653–0.724); 0% chance of being near-best, 0% chance of being best.

[Start free — the Retrieval Optimizer is included](https://cloud.goodmem.ai/login?loc=home-optimizer)

Beyond text

## Every page, ready for multimodal models

Real documents aren’t plain text. The answer often lives in a table grid, a chart, or a drawing — exactly what text extraction flattens away. GoodMem renders every page to an image at ingestion, even from Word, PowerPoint, and Excel, so a multimodal LLM can read the page the way a person would.

page image · 6-21

![A catalog page titled Contractor's Grapple Matching Guide: a grid of excavator models against grapple models, where a dot means the grapple fits and a shaded cell means it does not.](https://goodmem.ai/assets/grapple-matching-guide-Dl5aDNW3.webp)

1. Query

   “Does a G136 grapple fit a 321D excavator?”

2. GoodMem retrieves the page

   Extracted text

   `321D  •  •`

   Two dots — but under which grapples? The columns are gone.

   Page image

   The grid survives: the 321D row, and every column header above it.

3. A multimodal LLM reads the page

   Answer: **No.**

   The 321D takes the G120 or G126 on a B linkage. Both G136 columns are shaded “No Match” on that row.

- PDF
- DOCX
- PPTX
- XLSX

Page images for PDF and all three Office formats, Excel included — rendered in-process, in pure Java, with no LibreOffice in the deployment.

[Read how GoodMem captures pages](https://goodmem.ai/resources/ingestion-quality)

## Two agents ask for the salary bands. Only one gets them.

Permissions are checked before anything is searched. If a key can’t read a space, the request fails outright, and nothing leaks from the spaces it can read.

Two agents send the same question. The support agent’s key is scoped to the Help Center, so its request for the Compensation space is refused with HTTP 403 before any search runs. The HR assistant’s key is scoped to Compensation, so it gets the L5 salary band.

“What’s the salary band for an L5 engineer?”

support-agent

scoped to Help Center

403

Permission denied

The key can’t read Compensation, so nothing was searched.

hr-assistant

scoped to Compensation

comp-bands-2026.xlsx · p. 3

**Engineering salary bands**

| Engineering L4 | $148k – $176k |
| -------------- | ------------- |
| Engineering L5 | $182k – $214k |
| Engineering L6 | $221k – $262k |

**Stored in**

Your PostgreSQL with pgvector. No proprietary store to migrate out of.

**Runs in**

GoodMem Cloud, your VPC, or fully air-gapped.

**Logged**

Per your policy: who asked, with which key, what they asked, and what came back.

**Supply chain**

Signed SLSA Build L3 provenance · distroless, non-root server image.

**Certified**

ISO/IEC 27001.

[See how GoodMem compares on security](https://goodmem.ai/enterprise-security)

Cloud or local. Your choice.

## GoodMem keeps your pipeline portable

Use frontier APIs and the open model stack with the same GoodMem retrieval pipeline. Connect vLLM, TEI, Ollama, and OpenRouter alongside your hosted providers.

- OpenAI Hosted API
- AWS

  AWS Bedrock Managed models
- Ollama Local models

GoodMem

One retrieval pipeline

> “The ability to swap providers, mix local and cloud infrastructure, and keep the retrieval pipeline consistent across all of them is genuinely impressive.”

B. — a GoodMem customer

[Explore all integrations](https://goodmem.ai/integrations)

Flexible deployment

## Build agents on your own terms

- ### Self-hosted

  Production-ready, free for commercial use.

  Free

  [Self-host GoodMem](https://goodmem.ai/quick-start)

  What’s included:

  - Embed and ship anywhere, royalty‑free
  - Unlimited memories per node
  - gRPC API and SDKs in five languages
  - No Auto-Optimizer

- Fully managed

  ### GoodMem Cloud

  Continuously optimized for your data, ready in seconds.

  From $15/mo

  [Start free](https://cloud.goodmem.ai/login?loc=home-cta)

  What’s included:

  - Auto-Optimizer fine-tunes your retrieval models
  - We run, monitor, scale, and update it
  - Usage-based pricing, no surprise overages
  - 14-day free trial, no credit card

- ### Enterprise

  For OEMs, regulated industries, and service providers.

  Custom

  [Contact sales](https://goodmem.ai/contact-sales?interest=deployment)

  What’s included:

  - On-prem or in your VPC
  - Auto-Optimizer on your own infrastructure
  - Custom LLM training by PAIR
  - Source code access and a 99.99% SLA

[See detailed pricing](https://goodmem.ai/pricing)

terminal

\# Install goodmem

user\@goodmem :\~$ `curl -s "https://get.goodmem.ai" | bash`

SDKs

- [Python](https://pypi.org/project/goodmem/)
- [TypeScript](https://www.npmjs.com/package/@pairsystems/goodmem)
- [Java](https://central.sonatype.com/artifact/ai.pairsys/goodmem-java)
- [.NET](https://www.nuget.org/packages/Pairsystems.Goodmem.Client)
- [Go](https://gemfury.com/pairsys/go:fury.io%2Fpairsys%2Fgoodmem)