Market benchmarks / compute economics / market structure
The price of compute, tracked.
GPU rental markets, market-implied forward curves, chip economics, token prices, and the gigawatt buildout — one independent desk for the machines that make intelligence.
01Forward curves
What the market thinks compute will cost
Instrument-specific near anchors, monthly-average tenors through Aug ’27, and the shape of implied supply pressure ahead.
● Market-implied term structure
Implied forward curve
First point uses a weekly exact-time or labeled front-month monthly-average anchor; remaining points are monthly-average contracts — shown in $/GPU·hr or % vs anchor
Data — endpoints, license, and freshness behind Implied forward curve
Full forward-curve snapshot: instrument-specific anchors, monthly tenors, methodology identity, and per-point ladder evidence.
curl -s https://aicomputetracker.com/forward-curves.jsonSpreadsheet twin of /forward-curves.json: one row per tenor with unit-encoded per-chip columns, quality, and ladder evidence.
curl -s https://aicomputetracker.com/curves.csvNo API key, CORS open (Access-Control-Allow-Origin: *) — fetch it from a browser, a notebook, or a cron job.
Licensed CC BY 4.0 — attribution “AI Compute Tracker (aicomputetracker.com), CC BY 4.0”.
Market snapshot
Cite — copy a citation for Implied forward curve
Copy-ready citation
Cite as: AI Compute Tracker, “Implied forward curve”, market snapshot as of 2026-10-12, https://aicomputetracker.com/#curves, AI Compute Tracker (aicomputetracker.com), CC BY 4.0, accessed <date you visit>
Replace “<date you visit>” with your access date.
Embed — copy an iframe snippet for Implied forward curve
Copy-paste iframe
<div style="position:relative;width:100%;aspect-ratio:16/10;">
<iframe src="https://aicomputetracker.com/embed/forward-curve/" title="Implied forward curve — AI Compute Tracker" loading="lazy" style="position:absolute;inset:0;width:100%;height:100%;border:0;"></iframe>
</div>The frame keeps this chart’s own freshness stamp and CC BY 4.0 attribution; it updates when the site redeploys.
Kalshi event ladders referenced to Ornn. Raw implied median = the 50% bid/ask-mid crossing; monthly back points use two [1,2,1] smoothing passes from the third month onward. Hollow points are thin or interpolated.
How to read this curve
The shape is the story. B200, H200, H100, A100 and RTX 5090 ease lower out the strip through Aug ’27.
Replay the archived market-implied term structure.
Computed signals
The market in five numbers
Derived from the current datasets on every build — never hand-entered.
- B200 on-demand specialist median
- $8.19/GPU·hr
- +42% y/yBlackwell demand outran supply
- NVIDIA data-center revenue · FY27 Q2
- $89.0B
- +117% y/yNext-quarter guide $108B total
- Companywide capex · 2026E
- $803B
- Indicative · mixed basisMixed accounting bases; Oracle uses net cash capex
- Archived GPT-4-class floor
- $0.18/1M
- −99.5% since Mar ’23DeepSeek V4 Flash · Apr 2026 · blended 3:1
- RTX 4090 venue spread
- 5.6×
- specialist median ÷ marketplace floormedian $0.89 of 1 ÷ Salad $0.16/GPU·hr — venues sell different products
Today’s read · market snapshot OCT 12 2026 · 06:28 UTC
Compute is becoming a commodity — the fourth factor of production — and for the first time it trades like one.
A100 repriced +4% since Oct 11. Its curve is easing: the strip drifts lower into the back months. RTX 5090 is the anchor of the board: ~$0.64 through Aug ’27 on a easing curve.
The monthly-average anchor is $0.95; the Aug ’27 tenor implies $0.75.
The marketplace floor moved −3.9% over the latest three-quarter window to $1.73/GPU·hr.
The archived GPT-4-class floor through 2026-04 was about 210× cheaper than launch at $0.18 per million tokens. Current list prices are reviewed separately.
The Sep 28, 2026 notice cites pending regulatory review and gives no replacement date. Venue monitor
Desk log · latest entries
Full log →- Curve sessions1 forward-curve session archived · week of Oct 12, 2026
- Quote cutRTX 4090 marketplace floor cut −38.5%
- Quote cutRTX 5090 marketplace floor cut −19.4%
- Quote riseB200 specialist median rose +18.4%
Derived from the immutable archives on every build — no stored editorial event file.
02The Board
Every tracked chip, one index grid
Series-labeled reference list prices for all 16 tracked chips — with venue-class spreads, forward anchors where a market lists one, and archive deltas that self-activate as the daily record grows.
The Board · one reference price per chip
Every chip in the survey, led by its citable reference series — the specialist on-demand median, falling back (always labeled) to the hyperscaler on-demand floor, then the marketplace-access floor. Deltas compare the two latest archive dates; all figures are posted list/ask prices, not transactions. Manual survey 2026-10-11; automated rows through 2026-10-11; forward anchors from the OCT 12 2026 · 06:28 UTC market snapshot.
| Chip | Reference list price | Δ vs prior aggregate | Class spread | Forward anchor | Archive trend |
|---|---|---|---|---|---|
| B300 | $9.25/GPU·hrspecialist on-demand median | +21.3%vs Jul 23, 2026 | 1.1×spec median ÷ mkt floor marketplace-access floor $8.34/GPU·hr, specialist on-demand median $9.25/GPU·hr | no listed forward | 2 days · 07-23–10-11 |
| B200 | $8.19/GPU·hrspecialist on-demand median | +18.4%vs Jul 23, 2026 | 2.3×hyp floor ÷ mkt floor marketplace-access floor $6.25/GPU·hr, specialist on-demand median $8.19/GPU·hr, hyperscaler on-demand floor $14.24/GPU·hr | $7.78Oct ’26 monthly avg | 2 days · 07-23–10-11 |
| GB200 NVL72 | $10.50/GPU·hrspecialist on-demand median | 0.0%vs Jul 23, 2026 | single class — no spread to measure | no listed forward | 2 days · 07-23–10-11 |
| H200 | $5.29/GPU·hrspecialist on-demand median | +20.5%vs Jul 23, 2026 | 2.0×hyp floor ÷ mkt floor marketplace-access floor $3.93/GPU·hr, specialist on-demand median $5.29/GPU·hr, hyperscaler on-demand floor $7.91/GPU·hr | $5.44Oct ’26 monthly avg | 2 days · 07-23–10-11 |
| H100 | $3.99/GPU·hrspecialist on-demand median | +3.0%vs Jul 23, 2026 | 4.0×hyp floor ÷ mkt floor marketplace-access floor $1.73/GPU·hr, specialist on-demand median $3.99/GPU·hr, hyperscaler on-demand floor $6.88/GPU·hr | $2.84Oct ’26 monthly avg | 2 days · 07-23–10-11 |
| A100 80GB | $2.30/GPU·hrspecialist on-demand median | 0.0%vs Jul 23, 2026 | single class — no spread to measure | $0.95Oct ’26 monthly avg | 2 days · 07-23–10-11 |
| MI355X | $2.59/GPU·hrspecialist on-demand median | — first obs. | specialist on-demand median $2.59/GPU·hr, hyperscaler on-demand floor $8.60/GPU·hr | no listed forward | building history |
| MI300X | $3.72/GPU·hrspecialist on-demand median | +7.8%vs Jul 23, 2026 | specialist on-demand median $3.72/GPU·hr, hyperscaler on-demand floor $6.00/GPU·hr | no listed forward | 2 days · 07-23–10-11 |
| L40S | $1.50/GPU·hrspecialist on-demand median | 0.0%vs Jul 23, 2026 | single class — no spread to measure | no listed forward | 2 days · 07-23–10-11 |
| RTX 5090 | $1.19/GPU·hrspecialist on-demand median | +20.2%vs Jul 23, 2026 | 4.8×spec median ÷ mkt floor marketplace-access floor $0.25/GPU·hr, specialist on-demand median $1.19/GPU·hr | $0.64Oct ’26 monthly avg | 2 days · 07-23–10-11 |
| RTX 4090 | $0.89/GPU·hrspecialist on-demand median | +29.0%vs Jul 23, 2026 | 5.6×spec median ÷ mkt floor marketplace-access floor $0.16/GPU·hr, specialist on-demand median $0.89/GPU·hr | no listed forward | 2 days · 07-23–10-11 |
| V100 | no comparable series1 quote outside comparison policies | — | single class — no spread to measure | no listed forward | building history |
| TPU v6e | $2.70/chip·hrhyperscaler on-demand floor | 0.0%vs Jul 23, 2026 | single class — no spread to measure | no listed forward | 2 days · 07-23–10-11 |
| TPU v5e | $1.20/chip·hrhyperscaler on-demand floor | 0.0%vs Jul 23, 2026 | single class — no spread to measure | no listed forward | 2 days · 07-23–10-11 |
| TPU v5p | $4.20/chip·hrhyperscaler on-demand floor | — first obs. | single class — no spread to measure | no listed forward | building history |
| Trainium2 | no comparable series1 quote outside comparison policies | — | single class — no spread to measure | no listed forward | building history |
03GPU cloud
Renting an H100, 2023 → today
US$ per GPU-hour. From $12 at peak scarcity to a stable ~$4; primary on-demand now starts at $1.99, with separately labeled marketplace access near $1.73.
H100 SXM · provider and marketplace price
Quarterly midpoints by provider class
Data — endpoints, license, and freshness behind H100 SXM · provider and marketplace price
The complete GPU price-history ledger: every series and dated observation, with billing, comparison, quality, and raw-evidence fields.
curl -s https://aicomputetracker.com/price-history.jsonPer-accelerator shard of the price-history ledger — the same series and point shapes filtered to one chip.
curl -s https://aicomputetracker.com/price-history/h100.jsonNo API key, CORS open (Access-Control-Allow-Origin: *) — fetch it from a browser, a notebook, or a cron job.
Licensed CC BY 4.0 — attribution “AI Compute Tracker (aicomputetracker.com), CC BY 4.0”.
Live quotes
Cite — copy a citation for H100 SXM · provider and marketplace price
Copy-ready citation
Cite as: AI Compute Tracker, “H100 SXM · provider and marketplace price”, live quotes as of 2026-10-11, https://aicomputetracker.com/#gpu-cloud, AI Compute Tracker (aicomputetracker.com), CC BY 4.0, accessed <date you visit>
Replace “<date you visit>” with your access date.
Provider list prices, launch announcements, and archived snapshots; legacy quarterly midpoints ±15%. Current 2026-10-11 quote-table aggregates are provisional unless an evidence record pins an exact archived Git object.
04The macro picture
Compute demand is compounding
The frontier training-compute trend runs at 4.4× a year. Companywide capex guidance uses different lease, fiscal-year and cash definitions; the total is indicative.
Training compute of notable models
Total FLOP, log scale — estimates marked
Epoch AI estimates and lab disclosures; estimated points carry wide error bars.
Big-tech capital expenditure
US$B per calendar year (2026 = guidance)
Company disclosures, except Amazon’s secondary earnings transcript. Microsoft includes finance leases; Oracle FY27 uses net cash capex versus gross prior years and retains July guidance. Basis and dated evidence on the Buildout page.
05Inference
Intelligence, repriced monthly
Current provider prices reviewed 2026-10-11. The separate GPT-4-class history records a roughly 210× decline through Apr 2026.
Archived GPT-4-class floor
$0.18/1M
−99.5% since Mar ’23 · DeepSeek V4 Flash · Apr 2026
GPT-5.5 · current list price
$11.25/1M
+227% vs GPT-5 · Reviewed 2026-10-11; delta is vs archived GPT-5
Batch & cached reality
−50 / −90%
Batch APIs ~50% off list; cached input ~90% off — production pays below list
06Quick answers
The methodology, without the footnotes
The three definitions readers need before comparing the curves, provider classes, or token-price series.
How are the GPU forward curves derived?
From Kalshi event-contract ladders on GPU rental prices referenced to Ornn's hourly index. The first point uses a future exact-time weekly implied median when a usable ladder exists, otherwise an explicitly labeled front-month monthly-average implied median. Each raw implied level is the ladder's 50% crossing. The monthly strip receives two [1,2,1] smoothing passes from the third monthly tenor onward. Thin and interpolated points are labeled; none are current spot or executable quotes.
What does the blended token price mean?
Blended price = (3 × input + output) ÷ 4 per million tokens, reflecting a typical 3:1 read/write mix at standard list rates.
What are the provider classes?
Under primary on-demand policy primary-on-demand-v1, hyperscaler is the undiscounted big-cloud provider-list floor and specialist is the AI-native provider-list median. Marketplace access is the separately labeled open-market floor on community hardware.
07Go deeper
Explore the full desk
01
GPU Cloud
Primary on-demand rates for 16 chips across 19+ providers, with separately labeled marketplace access and history.
H100 on-demand from $1.99/hr
02
Hardware
Specs and economics for every accelerator that matters, V100 → Rubin.
Dense BF16 $/PFLOP·hr
03
Inference
Token prices across OpenAI, Anthropic, Google, xAI, and DeepSeek.
Archived Apr ’26 floor $0.18/1M
04
Buildout
Capex, NVIDIA’s run-rate, and every gigawatt-class cluster.
~36 GW frontier pipeline