AIForge Works - Weekly Pricing Pulse - May 18, 2026
About Pulse
The Weekly Pricing Pulse is what changed in published GPU cloud pricing across 9 providers in the last 7 days. Every week, AIForge Works runs a fresh harvest across AWS, Azure, GCP, OCI, CoreWeave, Lambda, Vultr, Nebius, and Crusoe, diffs the new data against the prior week, and surfaces the moves worth knowing about: biggest price changes on a per-GPU basis, new silicon that just appeared in any catalog, and SKUs that disappeared.
What the Pulse is NOT: it is not a live quote, not workload-specific advice, and not personalized to your account. The numbers reflect public catalog prices at the moment of harvest. For a live quote, check the provider. For workload-specific recommendations, use Cloud Advisor. To see neighboring SKUs, regions, or commitment terms for any mover that catches your eye, open the Multi-Cloud Pricing Analyzer. For an audit of your own GPU spend against a comparable floor, use GPU Audit.
Editor's note
This was a quiet launch week for Pulse. Comparable on-demand GPU pricing stayed flat over the 1% reporting threshold, and comparable 1-year and 3-year reserved or committed terms were effectively unchanged. The only visible catalog change was at Vultr, which added a few GPU entries and removed a handful. That is worth noting, but it is catalog churn rather than a broad pricing signal.
For procurement teams, this is a validation week rather than a repricing week. If you are refreshing a shortlist, recheck current catalog availability before carrying forward last week's assumptions, especially when evaluating Vultr regions. Keep committed-rate assumptions unchanged until reserved-term movement appears in the public catalogs.
This is the inaugural Pulse. Over the next few cycles, we will separate pricing movement from catalog movement more clearly, watch whether quiet committed terms stay quiet, and build a baseline for normal weekly drift across the nine-provider market. Feedback welcome at pulse@aiforge.works.
What moved this week
No provider had comparable on-demand price rows over the 1% movement threshold. Vultr changed in the catalog through new or removed entries, shown below.
- On-demand pricing was quiet this week. We did not find any public on-demand GPU price changes above the 1% reporting threshold; the visible movement is catalog availability.
Top on-demand movers by provider
No rows met the 1% movement threshold for this section.
Up to 5 rows per provider, ranked by absolute per-GPU rate change. See Methodology for how per-GPU rates are computed.
Provider roll-up
| Provider | Class | New SKUs | Deprecations | On-demand entries >1% |
|---|---|---|---|---|
| AWS | hyperscaler | 0 | 0 | 0 |
| Azure | hyperscaler | 0 | 0 | 0 |
| GCP | hyperscaler | 0 | 0 | 0 |
| OCI | hyperscaler | 0 | 0 | 0 |
| CoreWeave | neocloud | 0 | 0 | 0 |
| Crusoe | neocloud | 0 | 0 | 0 |
| Lambda | neocloud | 0 | 0 | 0 |
| Nebius | neocloud | 0 | 0 | 0 |
| Vultr | neocloud | 3 | 5 | 0 |
| Totals | - | 3 | 5 | 0 |
Movement counts are comparable provider + region + instance/SKU entries over the 1% movement threshold. They are row counts, not counts of unique instance types.
New silicon this week
- Vultr added
vbm-72c-480gb-gh200-gpuin Frankfurt, DE. H200 96 GB, 1-GPU pack. On-demand instance price: $1.9900/hr ($1.9900/hr per GPU). - Vultr added
vcg-a40-4c-20g-8vramin Frankfurt, DE. NVIDIA A40 48 GB, 1-GPU pack. On-demand instance price: $0.2880/hr ($0.2880/hr per GPU). - Vultr added
vbm-72c-480gb-gh200-gpuin New Jersey, US. H200 96 GB, 1-GPU pack. On-demand instance price: $1.9900/hr ($1.9900/hr per GPU).
Gone this week
- Vultr removed
vcg-a40-6c-30g-12vramfrom Bangalore, IN. Last seen on-demand: $0.4320/hr (per-GPU: $0.4320/hr). - Vultr removed
vbm-72c-480gb-gh200-gpufrom Market Unlisted (public feed). Last seen on-demand: $1.9900/hr (per-GPU: $1.9900/hr). - Vultr removed
vcg-a16-48c-496g-128vramfrom New Jersey, US. Last seen on-demand: $3.7670/hr (per-GPU: $0.4709/hr). - Vultr removed
vcg-a40-6c-30g-12vramfrom New Jersey, US. Last seen on-demand: $0.4320/hr (per-GPU: $0.4320/hr). - Vultr removed
vcg-a40-8c-40g-16vramfrom Sydney, AU. Last seen on-demand: $0.5750/hr (per-GPU: $0.5750/hr).
Methodology
What we compared. This Pulse compares last week's harvest (May 11, 2026) against this week's (May 18, 2026). Both harvests run automatically against the same 9 cloud providers.
How we compare prices fairly. Different providers package GPUs differently: some sell single-GPU instances, others sell 8-GPU instances. To compare apples-to-apples, we divide every instance price by the number of GPUs on that instance. All movement percentages in this Pulse are based on that per-GPU hourly rate, not the raw instance price.
What counts as a real move. We only flag price changes greater than 1%. Anything smaller is treated as rounding noise and excluded.
Reserved/committed terms. Comparable 1-year and 3-year reserved or committed price rows were checked separately. No comparable rows moved more than 1% this week, so no reserved/committed mover section is shown.
Currency. All prices are in US dollars. Rows missing a clean price or verified GPU specs are dropped before comparison.
How we choose summary callouts. The short summary only highlights price moves that are broad enough to read as a market signal. Narrow one-provider movements may still appear in detail tables, but we do not treat them as market-wide repricing.
This week's coverage. All 9 providers reported clean harvests. After normalizing 5.46M raw pricing rows against verified hardware specs, the broader MPA catalog for this harvest contains 108,974 all-in catalog instance rows. This Pulse narrows that catalog to roughly 14,000 comparable on-demand GPU offer keys for week-over-week movement analysis.
This Pulse compiles publicly published prices. It is not a live quote and is not workload-specific advice. See the About section at the top for links to AIForge Works pricing and availability tools.
Next Pulse: Week of 2026-05-25.