Overview
This is a repackaged open source software product wherein additional charges apply for independent packaging, configuration and setup documentation. Software Sushi packages Apache-2.0 Haystack and Hayhooks with a local document REST adapter, exact dependency locks and an operating guide. This product is independent of deepset and does not include deepset commercial platform features or upstream enterprise support. Dependencies retain their own licenses.
Store up to 1,000 local text documents in SQLite and query them through a Haystack BM25 pipeline. Results include source IDs, exact source text and retrieval scores. Upload up to 100 documents per request, each with up to 16,384 characters. Retrieval is local and needs no model. To generate answers, configure an OpenAI-compatible chat-completions model and credentials that you supply after launch. Questions and retrieved content are sent to your selected provider only when generation is requested. No model weights, provider credentials or inference allowance are included; provider charges and terms are separate. Inspect returned sources because model answers can be incorrect.
Launch t3.medium or larger x86_64 with at least 4 GiB RAM and a 20 GiB root volume. The enabled systemd service starts automatically. First boot generates a per-instance API access identity. Hayhooks listens on loopback port 1416 as an authenticated JSON REST API; there is no bundled browser UI. Keep inbound API ports closed and use SSH or Systems Manager Session Manager for access. Run sudo -iu ubuntu bash /opt/sushi/docs/demo.sh for a synthetic document task with exact-source verification and independent saved-file and API readback. Read /opt/sushi/docs/README.md for REST examples, provider configuration, persistence and recovery.
Documents persist under /var/lib/haystack across service restarts and instance reboots. This is a single-instance package using local disk and an in-memory retrieval index rebuilt per query; it does not provide replicated storage or durable background execution. Back up storage and configuration through your approved procedure. EC2, EBS, networking and any model-provider costs are separate from software charges.
Software costs USD 0.03 per running instance-hour on each offered type. Public English guide: https://softwaresushi-marketplace-branding-public.s3.us-east-1.amazonaws.com/docs/haystack/61d3f04c42027eb16ed9/getting-started.html
Highlights
- Haystack 3.4.1 and Hayhooks 2.1.0 with local SQLite document storage and BM25 retrieval returning exact source text.
- Per-instance authentication and loopback REST API on port 1416, with customer setup and recovery instructions.
- Keyless document walkthrough with independent readback; optional answer generation through a model you supply.
Details
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Cost/hour |
|---|---|
t3.medium Recommended | $0.03 |
t3.large | $0.03 |
t3.xlarge | $0.03 |
t3.2xlarge | $0.03 |
Vendor refund policy
We offer a full refund of software charges for the first 48 hours billing starts if the product does not perform as described. To request a refund, email info@softwaresushi.com with your EC2 instance ID and a description of the issue. AWS infrastructure costs (EC2, EBS, data transfer) are not eligible for refund by us.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
64-bit (x86) Amazon Machine Image (AMI)
Amazon Machine Image (AMI)
An AMI is a virtual image that provides the information required to launch an instance. Amazon EC2 (Elastic Compute Cloud) instances are virtual servers on which you can run your applications and workloads, offering varying combinations of CPU, memory, storage, and networking resources. You can launch as many instances from as many different AMIs as you need.
Version release notes
Haystack 3.4.1 and Hayhooks 2.1.0 on Ubuntu 24.04 with automatic service startup, authenticated document API, SQLite persistence, local BM25 retrieval and an English walkthrough. Optional answer generation uses an OpenAI-compatible model and credentials you supply after launch. Recommended t3.medium with at least 4 GiB RAM and 20 GiB encrypted root storage.
Additional details
Usage instructions
Launch t3.medium or larger x86_64 with at least 4 GiB RAM and 20 GiB encrypted gp3 root storage and require IMDSv2. Software is USD 0.03 per running instance-hour; EC2, EBS, networking and model-provider usage are separate. For Systems Manager Session Manager, attach an appropriate instance profile and provide outbound endpoint access; the session initially opens as ssm-user. Alternatively connect with SSH as ubuntu through your approved administrative network. Keep inbound application ports closed. Wait with sudo cloud-init status --wait, check sudo -iu ubuntu sushi-haystack status, and run sudo -iu ubuntu bash /opt/sushi/docs/demo.sh. The authenticated JSON REST API listens on loopback port 1416 and has no browser dashboard. The login banner prints the authenticated port-forward command. The demo stores synthetic local documents, retrieves exact sources and independently rereads saved files and the API. Results are saved under ~/haystack-results; documents persist in SQLite under /var/lib/haystack. Local retrieval requires no model. Optional generation sends your question and retrieved documents to an OpenAI-compatible chat-completions provider that you configure after launch. No provider key, model weights or inference allowance is bundled. Configure credentials in an unrecorded administration session using sudoedit, as described in /opt/sushi/docs/README.md. Never put credentials in command arguments or recordings. Configure backups and retention. Full public English guide: https://softwaresushi-marketplace-branding-public.s3.us-east-1.amazonaws.com/docs/haystack/61d3f04c42027eb16ed9/getting-started.html
Support
Vendor support
Software Sushi provides AMI packaging and setup guidance. Contact info@softwaresushi.com for installation and packaging questions. Upstream deepset enterprise support and hosted subscriptions are not included.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.