Sold by: Software Sushi
Deployed on AWS
This is a repackaged open source software product wherein additional charges apply for the packaged software. Run a baked Qwen2.5-0.5B-Instruct model through an authenticated LocalAI CPU API. Software: USD 0.03 per instance-hour; EC2, EBS and data transfer are separate.
Overview
Run a small instruction model locally and integrate its text responses through the OpenAI-compatible /v1/models and /v1/chat/completions endpoints. The package includes LocalAI, a CPU backend and the selected Qwen2.5-0.5B-Instruct Q4_K_M model, exposed as sushi-qwen. No post-launch model download or external inference provider account is required for the supplied local model.
Launch t3.large with 8 GiB RAM and at least 48 GiB encrypted gp3 root storage, require IMDSv2 and connect with Session Manager. Wait for automatic first boot, then run sudo -iu ubuntu bash /opt/sushi/docs/demo.sh. The included task checks unauthenticated rejection, model identity and a bounded arithmetic completion, retaining request and response files under the ubuntu user's localai-demo directory.
The authenticated API listens only on the instance loopback interface at port 8080. Use it from the instance or through a Session Manager tunnel. There is no web UI. Keep API port 8080 and backend port 50051 closed to inbound traffic. First boot creates the per-instance credential; the supplied client reads its mode-0600 configuration without printing it.
The configured service permits one concurrent backend inference, uses two CPU threads and a 2048-token context. The recommended launch profile provides headroom for the operating system, context, caches and application state; it is not a measured physical minimum or a throughput guarantee. Review generated text before using it in your application.
The software charge is USD 0.03 per running instance-hour. EC2, EBS and data transfer charges are separate. LocalAI is MIT-licensed; the selected Qwen model has separate Apache-2.0 terms. Read the installed notices and model card. Software Sushi packages this software independently and does not claim endorsement by the upstream projects or AWS.
Highlights
- Call a local Qwen2.5-0.5B-Instruct Q4_K_M CPU model as sushi-qwen through authenticated text-chat endpoints, without an external inference account.
- Use the included ubuntu demo to check model identity and an arithmetic completion, with original request and response files retained in a dedicated results directory.
- Launch t3.large with 8 GiB RAM and 48 GiB encrypted storage; connect through Session Manager while keeping application and backend ports closed to inbound traffic.
Details
Sold by
Categories
Delivery method
Delivery option
64-bit (x86) Amazon Machine Image (AMI)
Latest version
Operating system
Ubuntu 24.04
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Cost/hour |
|---|---|
t3.large Recommended | $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 billed by AWS and are not eligible for refund by us.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
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
LocalAI 4.12.0 on Ubuntu 24.04 x86_64 with an authenticated CPU model API and the bundled Qwen2.5-0.5B-Instruct Q4_K_M model exposed as sushi-qwen. Includes an ubuntu demo for model identity and a bounded arithmetic completion, retaining original request and response files. Verified version: 0.1.0 (LocalAI 4.12.0). Source AMI: ami-06e86b2de0b04f180.
Additional details
Usage instructions
- Launch Ubuntu 24.04 x86_64 on t3.large with 8 GiB RAM and at least 48 GiB encrypted gp3 root storage. Require IMDSv2. Attach your own Session Manager-capable instance profile and provide the Systems Manager connectivity required by your network. Keep inbound API port 8080 and backend port 50051 closed.
- Open a Session Manager shell and wait for cloud-init and automatic first boot. Run sudo -iu ubuntu bash /opt/sushi/docs/demo.sh. The task reads the generated per-instance credential without displaying it, checks unauthenticated rejection and the selected model, and sends a bounded arithmetic question to the actual local model. Original request and response files are saved under ~/localai-demo/ for ubuntu.
- Use the OpenAI-compatible API on instance loopback port 8080, with model sushi-qwen. This product has no web UI. Run clients on the instance or use a Session Manager port-forwarding session. The on-instance guide is /opt/sushi/docs/README.md. Keep ~/.config/localai/client.json confidential and do not paste its contents into command arguments, logs or support requests.
- The selected model is baked into the image and requires no post-launch model download or external inference provider account. The configuration permits one concurrent inference, two CPU threads and a 2048-token context. The launch profile is not a performance or minimum-hardware guarantee.
- Back up /etc/localai, /var/lib/localai and the ubuntu client configuration together. Restore matching identity and state together. Software costs USD 0.03 per running instance-hour; EC2, EBS and data transfer are billed separately. Guide: https://softwaresushi-marketplace-branding-public.s3.us-east-1.amazonaws.com/docs/localai/getting-started-93e67a6f937918e6.html .
Resources
Vendor resources
Support
Vendor support
Software Sushi provides AMI packaging and setup guidance. Contact info@softwaresushi.com for installation questions. Support coverage requires a separate agreement; this offer does not include upstream project support.
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.
Similar products
Run an Agno AgentOS API with a CSV calculation tool and persistent SQLite sessions on your own EC2 instance. This is paid repackaged open-source software.
Pre-installed and security-hardened OpenClaw AI agent AMI. Launch an EC2 instance, add your credentials, and start using your agent. Ready to configure out of the box.
This product has charges associated with it for seller support and packaging. Run the open-source Browser Use Python agent with local headless Chromium and your own model provider on Ubuntu 24.04.
Nanobot with localhost WebUI, native Python tools and a private Telegram channel. Customer cloud-model credentials are required after launch.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.