Ranking 01
Cheapest
Lowest true cost per GPU-hour, in stock or stock unknown, newest reading per provider
Theta EdgeCloud1× · 00:19 UTC$0.20
Lium1× · Brazil$0.22- 4
Infomaniak1× · Switzerland$0.24 - 5
gcube2× · Korea$0.26
NVIDIA · Turing · Datacenter · 37 providers list itNVIDIA T4 cloud prices
Rented today by 26 providers with a current on-demand offer, from $0.20 to $1.22 per GPU-hour, median $0.49. Every number below links to the page it was read from, with its read date.
Memory16GB
Bandwidth0.30TB/s
FP16 tensor65TFLOPS
FP328.1TFLOPS
Power70W
Specifications: manufacturer figures.
Cheapest on-demand · read 00:32 UTC
$0.20/ GPU·h
iwinv1× T4 · 16 GB · Korea- Median
- $0.49 26 providers, once each
- 30 days
- −13.3%
- 90 days
- −13.3%
Prices by rental mode · On-demand · Spot · Reserved · ServerlessT4 prices today
Cheapest, median and spread among current offers. Medians count one floor per marketplace and need at least three offers.
Cheapest$0.20iwinv
Median$0.49n = 87
Highest$1.22among current offers
Current offers13726 providers · last 45 days
Ever listed16931 providers on file
Where today's on-demand offers sitone dot per eligible offer (87) · the pool behind the tiles · read Nov 2, 2026 00:32 UTC
Provider offer84Marketplace floor3Median$0.49Middle half$0.50 – $0.94
87 eligible offers from 26 providers, binned every $0.05: from $0.20 (iwinv) to $1.22 (Naver Cloud). 18 further marketplace listings sit above their marketplace's floor and are not drawn (one floor per marketplace, as in the tiles). Hover a dot for the provider, price, configuration, region and read date; click it to open the provider.
On-demand price history · 34 daysdaily median across all providers · 2 provider lines from their 30-day series · read Nov 2, 2026
Median · all providers$0.49Providers
Today $0.4930 days -13.3%34 days -13.3%Low $0.10 · Oct 23High $0.63 · Oct 30
Daily median on-demand price per GPU-hour across all providers, 34 days to Nov 2, 2026. Provider lines: the provider's cheapest current on-demand offer, its listed price over the last 30 days (read Nov 2, 2026). Drag to zoom, double-click to reset.
Today's podium · on-demand · readings of the last 45 days · reference Nov 2, 2026 00:32 UTCWho rents the T4 best today
Cheapest true cost, best throughput per dollar, highest Trust Score. One provider per step, ten places, every row dated.
Ranking 02
Best value
Most throughput per dollar for LLM inference · Trust ≥ 60 first, then provisional · estimated
Google Cloud1× · Australia +24Value 43$0.35E2E Networks1× · IndiaValue 32$0.45
- 4
AWS1× · Germany +3Value 23$0.53 - 5
Baseten1× · 00:32 UTCValue 16$0.63
Ranking 03
Most trusted
Highest Trust Score among providers with an eligible offer · unscored providers follow, cheapest first
- Baseten1× · 00:32 UTCTrust 87$0.63
Microsoft Azure1× · Australia +17Trust 85$0.34
- 4Google Cloud1× · Australia +24Trust 81$0.35
- 5AWS1× · Germany +3Trust 79$0.53
87 eligible on-demand offers from 26 providers · median $0.49 (n = 87) · not ranked: 10 unavailable, 2 resold · readings of the last 45 days · reference Nov 2, 2026 · 00:32 UTCRankings are never sponsored. How we rank →
All offers · 376 readings on file · 37 providersEvery T4 offer, with its receipt
Current offers ranked by true cost, each with its read date. Older readings stay listed below, in grey, and never enter a ranking.
| Provider | Configuration | Region | Stock | List | True cost | Read | |
|---|---|---|---|---|---|---|---|
 Clore.aiTrust 49 | 4× T42 vCPU · 16 GB RAM · 31.307 GB disk | Brazil | Unavailable | $0.02 | $0.02 | 00:03 UTCfresh | Receipt |
| Clore.aiTrust 49 | 4× T42 vCPU · 16 GB RAM · 212.473 GB disk | Brazil | Unavailable | $0.02 | $0.02 | 00:03 UTCfresh | Receipt |
| Clore.aiTrust 49 | 4× T42 vCPU · 16 GB RAM · 33.344 GB disk | Brazil | Unavailable | $0.02 | $0.02 | 00:03 UTCfresh | Receipt |
| Clore.aiTrust 49 | 4× T42 vCPU · 16 GB RAM · 31.785 GB disk | Brazil | Unavailable | $0.02 | $0.02 | 00:03 UTCfresh | Receipt |
| Clore.aiTrust 49 | 4× T42 vCPU · 16 GB RAM · 31.762 GB disk | Brazil | Unavailable | $0.02 | $0.02 | 00:03 UTCfresh | Receipt |
| Clore.aiTrust 49 | 4× T42 vCPU · 16 GB RAM · 30.911 GB disk | Brazil | Unavailable | $0.02 | $0.02 | 00:03 UTCfresh | Receipt |
| Clore.aiTrust 49 | 4× T42 vCPU · 16 GB RAM · 31.026 GB disk | Brazil | Unavailable | $0.02 | $0.02 | 00:03 UTCfresh | Receipt |
| Clore.aiTrust 49 | 4× T42 vCPU · 16 GB RAM · 174.193 GB disk | Brazil | Unavailable | $0.02 | $0.02 | 00:03 UTCfresh | Receipt |
 Vast.aiTrust 78marketplace flooroutlier · suspended | 4× T432 vCPU · 53.7 GB RAM · 965.7 GB disk | Korea | In stock | $0.04 | $0.04 | 00:03 UTCfresh | Receipt |
| Vast.aiTrust 78 | 1× T48 vCPU · 13.4 GB RAM · 241.4 GB disk | Korea | In stock | $0.04 | $0.04 | 00:03 UTCfresh | Receipt |
| Vast.aiTrust 78 | 2× T416 vCPU · 26.9 GB RAM · 482.9 GB disk | Korea | In stock | $0.04 | $0.04 | 00:03 UTCfresh | Receipt |
| Vast.aiTrust 78 | 1× T48 vCPU · 23.4 GB RAM · 655 GB disk | US | In stock | $0.11 | $0.11 | 00:02 UTCfresh | Receipt |
| 1× T4 | Akash Network providers (decentralised) | Unavailable | $0.11 | $0.11 | 00:20 UTCfresh | Receipt | |
| Vast.aiTrust 78 | 1× T48 vCPU · 31.3 GB RAM · 225 GB disk | US | In stock | $0.12 | $0.12 | 00:03 UTCfresh | Receipt |
| Vast.aiTrust 78 | 2× T472 vCPU · 94.2 GB RAM · 1,400 GB disk | US | In stock | $0.12 | $0.12 | 00:02 UTCfresh | Receipt |
| Vast.aiTrust 78 | 1× T436 vCPU · 47.1 GB RAM · 700 GB disk | US | In stock | $0.12 | $0.12 | 00:02 UTCfresh | Receipt |
| Vast.aiTrust 78 | 4× T456 vCPU · 503.5 GB RAM · 725.8 GB disk | Czechia | In stock | $0.15 | $0.15 | 00:00 UTCfresh | Receipt |
| Vast.aiTrust 78 | 2× T418.666666666666664 vCPU · 62.8 GB RAM · 303.1 GB disk | Czechia | In stock | $0.15 | $0.15 | 00:00 UTCfresh | Receipt |
| Vast.aiTrust 78 | 2× T428 vCPU · 251.8 GB RAM · 362.9 GB disk | Czechia | In stock | $0.15 | $0.15 | 00:00 UTCfresh | Receipt |
| Vast.aiTrust 78 | 1× T424 vCPU · 15.4 GB RAM · 135.5 GB disk | Vietnam | In stock | $0.15 | $0.15 | 00:01 UTCfresh | Receipt |
| Vast.aiTrust 78 | 1× T49.333333333333332 vCPU · 31.4 GB RAM · 151.6 GB disk | Czechia | In stock | $0.15 | $0.15 | 00:00 UTCfresh | Receipt |
| Vast.aiTrust 78 | 1× T416 vCPU · 108.1 GB RAM · 404 GB disk | Canada | In stock | $0.16 | $0.16 | 00:00 UTCfresh | Receipt |
| iwinv | 1× T46 vCPU · 60 GB RAM | Korea | Stock unknown | $0.20 | $0.20 | 00:31 UTCfresh | Receipt |
| iwinv | 2× T412 vCPU · 120 GB RAM | Korea | Stock unknown | $0.20 | $0.20 | 00:31 UTCfresh | Receipt |
| iwinv | 4× T424 vCPU · 240 GB RAM | Korea | Stock unknown | $0.20 | $0.20 | 00:31 UTCfresh | Receipt |
137 fresh offers · the 25 cheapest shown here, all of them in the comparison tool · 32 older readings of the same provider replaced by a newer one · prices per GPU-hour · true cost includes the add-ons the provider publishesOpen in the comparison tool →
Value · what the hour buysPrice per unit, and what fits
The same hour, divided by what the card gives you; then the smallest real configuration for the most downloaded open models.
| Unit | At the cheapest | At the median |
|---|---|---|
| per GB of VRAM · hour16 GB | $0.012 | $0.031 |
| per PFLOP FP16 · hour65 TFLOPS | $3.005 | $7.580 |
| per TB/s of bandwidth · hour0.30 TB/s | $0.651 | $1.642 |
| Model | Needs | Smallest config | $ / month |
|---|---|---|---|
| all MiniLM L6 v2Sentence Transformers | 1 GBFP8 | 1× T4at Vast.ai | $29 |
| paraphrase multilingual MiniLM L12 v2Sentence Transformers | 1 GBFP8 | 1× T4at Vast.ai | $29 |
| bge-m3BAAI | 1 GBFP8 | 1× T4at Vast.ai | $29 |
| Qwen3 0.6BQwen | 1 GBFP8 | 1× T4at Vast.ai | $29 |
| all-mpnet-base-v2Sentence Transformers | 1 GBFP8 | 1× T4at Vast.ai | $29 |
| bge-reranker-v2-m3BAAI | 1 GBFP8 | 1× T4at Vast.ai | $29 |
Monthly = 730 hours at the cheapest current on-demand price for that GPU count · models sized at FP8 when available, weights plus runtime overheadSize any model →
T4 · comparisons and collectionsCompare the T4 side by side
Side by side
Collections
Questions about renting the T4
How much does the T4 cost per hour?
Today the T4 rents from $0.20 per GPU-hour on-demand, with a median of $0.49 across 26 providers, each counted once at its cheapest current offer (87 current offers in all, readings of the last 45 days). Prices depend on the provider, the number of GPUs in the machine, the region and whether you rent on-demand, spot or reserved; true cost adds the mandatory add-ons each provider publishes.
Which cloud providers offer the T4?
29 providers have a current T4 offer in stock. Cheapest first: Vast Ai ($0.04), Iwinv ($0.20), Theta Edgecloud ($0.20), Lium ($0.22), Infomaniak ($0.24), Gcube ($0.26). Older readings from other providers stay listed on this page, in grey, and are never ranked.
How much memory does the T4 have?
16 GB of memory at 0.30 TB/s. At FP8 that fits all MiniLM L6 v2 (1 GB), paraphrase multilingual MiniLM L12 v2 (1 GB), bge-m3 (1 GB), Qwen3 0.6B (1 GB), all-mpnet-base-v2 (1 GB) on a single card; larger models need two to eight GPUs or a different card.
Where is the T4 cheapest right now?
The lowest true cost among eligible offers is $0.20 per GPU-hour at iwinv for a 1-GPU machine in Korea, read within the ranking window. Offers marked unavailable, older than 45 days, superseded by a newer reading, resold or outside the outlier band are listed but not ranked.
Is the T4 cheaper on spot or reserved?
Spot (interruptible) capacity starts at $0.07 per GPU-hour; reserved terms start at $0.17. Spot suits jobs that checkpoint; reserved suits machines that run all month.