Sold by: Multiverse Computing
Deployed on AWS
Hypernova-60B v2602 is an enhanced version of the Hypernova-60B model, introducing improved tool-calling capabilities built on a compressed GPT-OSS-120B derived architecture. It maintains frontier-level reasoning performance while operating at roughly half the model size through targeted knowledge distillation. The update primarily focuses on stronger agentic workflows and reliable tool interaction.
Overview
Hypernova-60B v2602 is an enhanced version of the Hypernova-60B model, introducing improved tool-calling capabilities built on a compressed GPT-OSS-120B derived architecture. It maintains frontier-level reasoning performance while operating at roughly half the model size through targeted knowledge distillation. The update primarily focuses on stronger agentic workflows and reliable tool interaction. In comparison to GPT-OSS-120B inference performance, Hypernova-60B v2602 achieves-
Highlights
- +39.5% higher throughput and -36% lower P99 latency
- Up to 50.8% faster Time-to-First-Token (TTFT)
- ~45% reduction in per-token generation time (TPOT & ITL)
Details
Sold by
Categories
Delivery method
Latest version
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 | Description | Cost/host/hour |
|---|---|---|
ml.g5.48xlarge Inference (Batch) Recommended | Model inference on the ml.g5.48xlarge instance type, batch mode | $3.06 |
ml.g5.48xlarge Inference (Real-Time) Recommended | Model inference on the ml.g5.48xlarge instance type, real-time mode | $3.06 |
ml.g6.48xlarge Inference (Batch) | Model inference on the ml.g6.48xlarge instance type, batch mode | $2.73 |
ml.g6.48xlarge Inference (Real-Time) | Model inference on the ml.g6.48xlarge instance type, real-time mode | $2.73 |
ml.p5.48xlarge Inference (Real-Time) | Model inference on the ml.p5.48xlarge instance type, real-time mode | $8.21 |
ml.p5e.48xlarge Inference (Real-Time) | Model inference on the ml.p5e.48xlarge instance type, real-time mode | $14.78 |
Dimensions summary
You pay by the hour for each host running model inference, so cost scales with runtime and the instance you pick. Pricing splits two ways. First, by inference mode: batch processes grouped requests, while real-time serves live responses. Second, by instance type, which reflects the underlying hardware size. The ml.g5.48xlarge and ml.g6.48xlarge are available in both batch and real-time modes. The ml.p5.48xlarge and ml.p5e.48xlarge run in real-time mode only. Choose the instance and mode that match your workload. There is no upfront commitment; you are billed per host hour used.
Top-of-mind questions for buyers
One host hour is one running instance active for one hour. You are billed per instance, per hour of runtime. Each instance type meters independently. If you run two instances at once, you accrue two host hours per clock hour. Partial hours follow standard metering for the instance.
Batch mode processes grouped requests together and bills per host hour while the batch job runs. Real-time mode keeps a host running to serve live requests and bills per host hour it stays active. Batch suits scheduled, high-volume jobs. Real-time suits interactive, on-demand responses.
Charges apply per host hour while the instance stays active, even if it serves no requests during that time. A real-time endpoint keeps the host running to accept live traffic. To stop software charges, shut down the instance. Underlying AWS infrastructure fees may still apply separately.
multiversecomputing.com
Helpful?
Vendor refund policy
Placeholder
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
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
CompactifAI Hypernova-60B 2602 inference on a vLLM engine
Additional details
Inputs
- Summary
Our model accepts an input compatible with openAI's chat completions endpoint.
- Input MIME type
- application/json
{ "model": "hypernova-60b", "messages": [ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Tell me about quantum computing."} ] }
{ "model": "hypernova-60b", "messages": [ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Tell me about quantum computing."} ] }
Support
Vendor 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
Reduce inference costs and increase inference speed by using Multiverse Computing's CompactifAI HyperNova 60B. In direct benchmarks against Mistral 3 Large, HyperNova 60B delivers outstanding gains:
This Ubuntu 26.04 Latest image has charges associated with it for security hardening and optimization provided by ProComputers. The image is automatically updated at launch with the newest security patches, so every instance starts on the most secure version available. Access is through the 'ubuntu' account using SSH public key authentication. Cloud-init takes care of initial configuration, ENA enables accelerated networking, and the root filesystem resizes itself at boot when you attach a larger volume. Based on Ubuntu 26.04 LTS, the system offers long-term support and a dependable environment for web servers, application back ends, and databases. Its small footprint and fast startup make Ubuntu 26.04 Latest a practical choice for LAMP stacks, container hosts, Python and PHP applications, and data stores such as MySQL, PostgreSQL, and MongoDB.
This is a Image Classification model from TensorFlow Hub
This is a Image Classification model from TensorFlow Hub
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.