Fastly Image Optimizer

Overview

Fastly Image Optimizer is a cloud service that transforms and delivers images at the edge for image-heavy sites and apps. It is best suited to teams that need many image variants without storing each one separately. Its appeal is configurable, cached processing; its cost case depends on request volume, since every request to the service counts toward billing.

Key features

Image Optimizer can resize, crop or trim, adjust quality, change orientation and convert images when requested. Developers configure these choices with query parameters and settings, avoiding the need to pre-generate and store separate versions. That is useful for sites with varied image placements, though teams with a small, fixed set of assets may have less need for request-time transformation.

Supported formats are WebP, AVIF, JPEG and PNG. Fastly also provides configurable defaults for WebP and JPEG quality, resizing filters, upscaling, animated GIF-to-video conversion and eligible file types. These defaults offer control over common processing behavior, while still requiring teams to choose settings that fit their output needs.

Processed images are cached and reused, allowing subsequent requests to be served with minimal latency. The service integrates with existing asset pipelines and CMS platforms, so teams can continue using their current image sources. It can be enabled through the Fastly control panel or API; Compute SDK access is through a beta program. Requests through the SDK, like other Image Optimizer requests, are billable.

Fastly says its global edge network is designed for high-throughput delivery to media sites, ecommerce platforms and SaaS applications. It also states that it is ISO/IEC 27001:2022 certified, audited against SOC 2 Type 2 and key GDPR requirements, PCI DSS Level 1 compliant, and audited against relevant HIPAA sections. Those controls may matter to organizations with compliance requirements, but do not remove the need to assess their own obligations.

Pricing

Image Optimizer has a free allowance of 100,000 image requests per month at free. Above that, the published volume tiers are billed per 10,000 image requests: $0.50 a month for 100,000–10 million requests; $0.15 a month for 10 million–50 million; $0.07 a month for 50 million–150 million; $0.03 a month for 150 million–500 million; $0.02 a month for 500 million–1 billion; $0.02 a month for 1 billion–5 billion; and $0.01 a month beyond 5 billion. Pricing is on request, so these tier figures should not be treated as a complete quote.

The free tier suits evaluation or low-volume use, but its monthly cap is quickly outgrown by high-traffic services. The lower paid-volume tiers cost more per 10,000 requests than the largest tiers, making accurate traffic estimates important. Count all Image Optimizer requests, including SDK traffic, when assessing the meter. Fastly's paid-from figure is 1500 /mo, so confirm the applicable total for a deployment rather than assuming the per-request tier alone represents the bill.

Platforms

Fastly Image Optimizer is a cloud service with API, Linux and web platform support. Fastly's broader platform also provides API access and supports deployment through the Fastly CLI, control panel, GitHub Actions, Terraform and APIs. The wider Fastly offering includes image optimization, origin shield, video delivery, edge compute and real-time logs; WAF is not included.

Who it's for

This is a strong fit for media, ecommerce and SaaS teams that serve many image sizes or formats and can benefit from edge caching. Teams already using Fastly can keep their existing image sources and configure transformations through familiar pipeline or CMS workflows. It is less compelling for low-traffic sites with few static assets, or for buyers who need predictable costs without first estimating image request volume.

Pros and cons

Pro: Request-time resizing, cropping, quality changes and format conversion avoid maintaining a separate stored variant for each use.

Pro: Cached results can reduce latency on repeat requests, while support for WebP, AVIF, JPEG and PNG covers common delivery formats.

Pro: Integration with existing pipelines and CMS platforms lets teams retain their current image sources.

Con: All requests count toward billing, including SDK requests, so usage can be difficult to budget without a reliable traffic estimate.

Con: Compute SDK access is in beta, which may not suit teams that require a generally available SDK workflow.

Con: The free allowance is capped at 100,000 requests per month, and the paid-from figure of 1500 /mo warrants checking the full quote before committing.

Alternatives

For broader image compression options, compare products in Image Compression Software; for edge delivery platforms, see CDN Software. Buyers focused on DDoS defenses can browse DDoS Protection Software, while teams evaluating application-layer defenses can consult Web Application Firewall Software.

F5 BIG-IP APM is a paid alternative with a free trial and broad platform support, including self-hosted deployments. Gcore DDoS Protection is worth considering for a freemium DDoS option with a free PAYG tier and API and web support. A10 Thunder ADC is a paid alternative with a free trial and API, Linux and self-hosted support. Andrisoft Wanguard offers a paid, trial-available option with API, Linux, self-hosted and web support; its Sensor license costs $595 a year.

AWS Shield is a freemium DDoS option with a free Standard tier for common network and transport layer protection configured automatically for AWS customers. Check Point DDoS Protector is a paid option for Linux, self-hosted and web deployments. Oracle Cloud Infrastructure Secret Management is a free API and web service for managing secrets. Cloudflare DDoS Protection offers a free plan with unmetered DDoS protection, a WAF and a free managed ruleset.

Verdict

Choose Fastly Image Optimizer if your image-heavy site needs configurable variants at the edge and you can forecast request volume closely enough to manage the meter. Its strongest reason to buy is the combination of on-demand transformations, cached reuse and compatibility with existing image workflows. Look elsewhere if your use is low-volume or cost certainty matters more than avoiding stored variants.