This month is still in progress. Figures update as prices change and are final on the 1st of next month.
AI API & GPU price report: November 2026
In November 2026 we recorded 369 price changes across AI model APIs and GPU clouds: 62 API price drops, 203 increases and 104 GPU rental price changes. The biggest cut was DeepSeek V4 Flash on OpenRouter (−84%).
GPU price index
On-demand price per GPU-hour across the GPU clouds we track, today, compared with the start of the month.
| GPU | Lowest | Cheapest on | Median | Median vs start | Offers |
|---|---|---|---|---|---|
| B300 | $7.10 | Modal | $8.99 | −2% | 13 |
| B200 | $3.50 | Vultr | $8.50 | +4% | 18 |
| H200 | $3.00 | Vast.ai | $5.29 | +3% | 21 |
| H100 | $1.67 | Vast.ai | $3.79 | +8% | 39 |
| A100 80GB | $0.52 | Vast.ai | $2.19 | +22% | 33 |
| L40S | $0.47 | Vast.ai | $1.56 | +0.3% | 14 |
| RTX PRO 6000 | $0.80 | Vast.ai | $2.58 | +5% | 17 |
| RTX 5090 | $0.37 | Vast.ai | $0.91 | +22% | 4 |
| RTX 4090 | $0.27 | Vast.ai | $0.64 | no change | 7 |
| MI300X | $1.85 | Vultr | $3.45 | no change | 5 |
Biggest API price drops
Price drops(62)
- DeepSeek V4 FlashonOpenRouterinput $0.09 → $0.0146/1M, output $0.18 → $0.0291/1M↓−84%
- DeepSeek V4 Pro (0813)onOpenRouterinput $0.66 → $0.132/1M, output $1.98 → $0.396/1M↓−80%
- Qwen 3 TTSonRuncrateper 1M chars $105 → $22.05/1M chars↓−79%
- Llama 3.1 8B InstructonWaveSpeed AIinput $0.22 → $0.05/1M, output $0.22 → $0.08/1M↓−77%
- Gemma 4 26B A4BonOpenRouteroutput $0.22 → $0.052/1M↓−76%
- DeepSeek V3.2onWaveSpeed AIinput $1 → $0.29/1M, output $1 → $0.43/1M↓−71%
- Qwen3 Coder 480B A35BonWaveSpeed AIinput $0.975 → $0.35/1M, output $4.875 → $1.5/1M↓−69%
- GLM 5.2onOpenRouterinput $0.5625 → $0.189/1M, output $1.8 → $0.594/1M↓−67%
- gpt-oss-20bonOpenRouteroutput $0.09 → $0.03/1M↓−67%
- Grok AuroraonxAIper image $0.05 → $0.02/image↓−60%
…and 52 more; see the daily pages under price changes.
Biggest API price increases
Price increases(203)
- MiMo V2.5onOpenRouterinput $0.119 → $0.4/1M, output $0.238 → $2/1M↑+740%
- DeepSeek V4 FlashonOpenRouterinput $0.0147 → $0.072/1M, output $0.0294 → $0.144/1M↑+390%
- DeepSeek V4 Pro (0813)onOpenRouterinput $0.132 → $0.5808/1M, output $0.396 → $1.742/1M↑+340%
- DeepSeek V4 ProonOpenRouterinput $0.2079 → $0.9/1M, output $0.4157 → $1.5/1M↑+333%
- Qwen3.8 27BonOpenRouterinput $0.045 → $0.04/1M, output $0.32 → $1.35/1M↑+322%
- Qwen3 Next 80B A3B InstructonInfronoutput $0.3444 → $1.1/1M↑+219%
- Gemini 3 ProonReplicateoutput $4 → $12/1M↑+200%
- Gemini 3.1 ProonReplicateoutput $4 → $12/1M↑+200%
- Gemma 4 26B A4BonOpenRouterinput $0.017 → $0.04/1M, output $0.052 → $0.15/1M↑+188%
- MiniMax M2.7onOpenRouterinput $0.21 → $0.6/1M, output $0.84 → $1.8/1M↑+186%
GPU rental price changes
GPU rental prices(104)
- V100 32GBonRuncrateon-demand $3.04 → $0.4/hr↓−87%
- L4onVast.aion-demand $0.25 → $0.1/hr↓−60%
- L4onRuncrateon-demand $1.24 → $0.58/hr↓−53%
- RTX PRO 5000onVast.aion-demand $0.6 → $0.37/hr↓−38%
- B300onRuncrateon-demand $14.3 → $8.93/hr↓−38%
- H200onVast.aion-demand $2.63 → $1.71/hr↓−35%
- V100 16GBonRuncrateon-demand $0.55 → $0.36/hr↓−35%
- H200 SXM5onOblivuson-demand $5.5 → $3.99/hr, reserved $4.47 → $2.99/hr↓−33%
- RTX A6000onRuncrateon-demand $0.74 → $0.57/hr↓−23%
- RTX 6000 AdaonRuncrateon-demand $1.21 → $0.96/hr↓−21%
Most active providers
- OpenRouter126 changes
- Runcrate90 changes
- Infron36 changes
- Vast.ai25 changes
- WaveSpeed AI20 changes
- RunPod13 changes
- Verda12 changes
- Google Cloud8 changes
- Together AI5 changes
- Lambda Cloud5 changes
Using these numbers
You're welcome to quote this report. Please link to it as the source:
Source: Inference Hub, AI API & GPU Price Report, November 2026 (https://inferencehub.org/price-report/2026-11)
How we collect prices: from each provider's own pricing page or API, checked weekly; see our methodology. Percentages compare the new price with the old; when input and output prices both moved, the larger move is shown. Get daily alerts by RSS or on Telegram.