SF·SIX / PRICE TRACKER
AI API price changes
Every recorded change of a published AI API price in the last 30 days: cuts, increases and new priced listings, by model and provider, newest first.
In the 7 days to 2026-10-17 (UTC), SF·SIX recorded 523 API price cuts and 366 price increases across 40 models, and 287 new priced listings. Over 30 days: 523 cuts, 366 increases. The largest cut: DeepSeek V4.1 Flash on Baidu (fp8), $0.30 / $1.20 to $0.0357 / $0.1428 (−88.1%).
Weekly summary
| Week of (UTC) | Price cuts | Price increases | New priced listings | Models with a price move |
|---|---|---|---|---|
| 2026-10-12 | 523 | 366 | 287 | 40 |
| 2026-10-05 | 0 | 0 | 0 | 0 |
| 2026-09-28 | 0 | 0 | 0 | 0 |
| 2026-09-21 | 0 | 0 | 0 | 0 |
| 2026-09-14 | 0 | 0 | 0 | 0 |
Counts per ISO week (Monday to Sunday, UTC). A model with several provider endpoints can move more than once.
Price cuts
| Date | Model | Provider and terms | Before | After | Change |
|---|---|---|---|---|---|
| 2026-10-1712:12 UTC | Kimi K3Moonshot AI | WaferUS regionmxfp4 | $0.50 / $14.90 | $0.36 / $14.90 | −2.6%in −28.0% |
| 2026-10-1711:56 UTC | GLM 5.3Z.ai (Zhipu) | Relace | $0.069 / $19.90 | $0.0632 / $4.40 | −77.2%in −8.4% · out −77.9% |
| 2026-10-1711:56 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Decartfp4 | $0.129 / $0.43 | $0.0825 / $0.275 | −36.0%in −36.0% · out −36.0% |
| 2026-10-1711:40 UTC | DeepSeek V4.1 FlashDeepSeek | Morphfp8 | $0.09 / $0.55 | $0.088 / $0.55 | −0.7%in −2.2% |
| 2026-10-1711:24 UTC | DeepSeek V4 Flash 0423DeepSeek | Baidufp8 | $0.0214 / $0.0428 | $0.0181 / $0.0361 | −15.7%in −15.7% · out −15.7% |
| 2026-10-1710:52 UTC | DeepSeek V4 Flash 0731DeepSeek | Phala | $0.44 / $1.32 | $0.308 / $0.924 | −30.0%in −30.0% · out −30.0% |
| 2026-10-1710:52 UTC | DeepSeek V4 Flash 0731DeepSeek | Relacefp4 | $0.0087 / $0.2829 | $0.0071 / $0.2829 | −1.6%in −18.4% |
| 2026-10-1710:52 UTC | DeepSeek V4 Pro 0813DeepSeek | Phala | $1.45 / $4.36 | $0.957 / $2.88 | −34.0%in −34.0% · out −34.0% |
| 2026-10-1710:36 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Relace | $0.0293 / $0.4182 | $0.0238 / $0.3966 | −7.5%in −18.8% · out −5.2% |
| 2026-10-1710:36 UTC | GLM 5.3 FlashZ.ai (Zhipu) | OpenInferencefp4 | $0.0293 / $0.4182 | $0.0238 / $0.3966 | −7.5%in −18.8% · out −5.2% |
| 2026-10-1710:36 UTC | DeepSeek V4.1 FlashDeepSeek | Relace | $0.0188 / $0.6246 | $0.0153 / $0.594 | −6.0%in −18.6% · out −4.9% |
| 2026-10-1710:36 UTC | DeepSeek V4.1 FlashDeepSeek | OpenInferencefp4 | $0.0198 / $0.6246 | $0.0153 / $0.594 | −6.4%in −22.7% · out −4.9% |
| 2026-10-1709:48 UTC | Kimi K3Moonshot AI | Relacefp4 | $0.27 / $14.23 | $0.27 / $13.98 | −1.7%out −1.8% |
| 2026-10-1709:48 UTC | DeepSeek V4 Flash 0731DeepSeek | OpenInferencefp4 | $0.0119 / $0.2971 | $0.0092 / $0.2971 | −2.4%in −22.7% |
| 2026-10-1709:48 UTC | DeepSeek V4 Flash 0731DeepSeek | Relacefp4 | $0.0119 / $0.2971 | $0.0087 / $0.2829 | −7.2%in −26.9% · out −4.8% |
| 2026-10-1709:48 UTC | DeepSeek V4 Pro 0813DeepSeek | Baidufp8 | $0.1544 / $0.4633 | $0.1333 / $0.40 | −13.7%in −13.7% · out −13.7% |
| 2026-10-1709:48 UTC | DeepSeek V4 Pro 0813DeepSeek | Relacefp4 | $0.155 / $0.5588 | $0.1233 / $0.3831 | −26.4%in −20.4% · out −31.4% |
| 2026-10-1709:48 UTC | Seedance 2.5 US Text to VideoByteDance Seed | fal | $0.0257 per 1000 tokens | $0.0214 per 1000 tokens | −16.7% |
| 2026-10-1709:48 UTC | Seedance 2.5 US Reference to VideoByteDance Seed | fal | $0.0257 per 1000 tokens | $0.0214 per 1000 tokens | −16.7% |
| 2026-10-1709:48 UTC | Seedance 2.5 US Image to VideoByteDance Seed | fal | $0.0257 per 1000 tokens | $0.0214 per 1000 tokens | −16.7% |
| 2026-10-1709:48 UTC | Seedance 2.0 US Text to VideoByteDance Seed | fal | $0.0168 per 1000 tokens | $0.014 per 1000 tokens | −16.7% |
| 2026-10-1709:48 UTC | Seedance 2.0 US Reference to VideoByteDance Seed | fal | $0.0168 per 1000 tokens | $0.014 per 1000 tokens | −16.7% |
| 2026-10-1709:48 UTC | Seedance 2.0 US Image to VideoByteDance Seed | fal | $0.0168 per 1000 tokens | $0.014 per 1000 tokens | −16.7% |
| 2026-10-1709:29 UTC | Qwen3.8 27BQwen (Alibaba) | Waferfp8 | $0.08 / $3.00 | $0.04 / $2.30 | −25.3%in −50.0% · out −23.3% |
| 2026-10-1709:29 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Relace | $0.0359 / $0.4182 | $0.0293 / $0.4182 | −3.8%in −18.4% |
| 2026-10-1709:29 UTC | GLM 5.3 FlashZ.ai (Zhipu) | OpenInferencefp4 | $0.0378 / $0.4182 | $0.0293 / $0.4182 | −4.8%in −22.5% |
| 2026-10-1709:13 UTC | DeepSeek V4.1 FlashDeepSeek | OpenInferencefp4 | $0.0256 / $0.6246 | $0.0198 / $0.6246 | −2.5%in −22.7% |
| 2026-10-1709:13 UTC | DeepSeek V4.1 FlashDeepSeek | Relace | $0.0243 / $0.6246 | $0.0188 / $0.6246 | −2.4%in −22.6% |
| 2026-10-1708:41 UTC | DeepSeek V4 Flash 0423DeepSeek | Baidufp8 | $0.0227 / $0.0454 | $0.0214 / $0.0428 | −5.6%in −5.6% · out −5.6% |
| 2026-10-1708:25 UTC | Kimi K3Moonshot AI | Morphfp8 | $0.51 / $14.90 | $0.30 / $14.90 | −3.8%in −41.2% |
| 2026-10-1708:25 UTC | DeepSeek V4 Flash 0731DeepSeek | OpenInferencefp4 | $0.0154 / $0.9653 | $0.0119 / $0.2971 | −67.1%in −22.7% · out −69.2% |
| 2026-10-1708:25 UTC | DeepSeek V4 Flash 0731DeepSeek | Relacefp4 | $0.0154 / $0.9653 | $0.0119 / $0.2971 | −67.1%in −22.7% · out −69.2% |
| 2026-10-1708:25 UTC | DeepSeek V4 Pro 0813DeepSeek | Wafermxfp4 | $0.25 / $5.00 | $0.25 / $2.50 | −43.5%out −50.0% |
| 2026-10-1708:25 UTC | DeepSeek V4 Pro 0813DeepSeek | Relacefp4 | $0.1972 / $0.7497 | $0.155 / $0.5588 | −23.7%in −21.4% · out −25.5% |
| 2026-10-1708:25 UTC | DeepSeek V4 Pro 0813DeepSeek | Baidufp8 | $0.2072 / $0.6217 | $0.1544 / $0.4633 | −25.5%in −25.5% · out −25.5% |
| 2026-10-1708:09 UTC | Qwen3.8 27BQwen (Alibaba) | Waferfp8 | $0.20 / $6.00 | $0.08 / $3.00 | −50.9%in −60.0% · out −50.0% |
| 2026-10-1708:09 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Morphfp8 | $0.069 / $0.65 | $0.066 / $0.65 | −1.1%in −4.3% |
| 2026-10-1708:09 UTC | GLM 5.3 FlashZ.ai (Zhipu) | OpenInferencefp4 | $0.0464 / $0.5578 | $0.0378 / $0.4182 | −23.7%in −18.5% · out −25.0% |
| 2026-10-1708:09 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Relace | $0.0464 / $0.5578 | $0.0359 / $0.4182 | −24.5%in −22.6% · out −25.0% |
| 2026-10-1707:37 UTC | DeepSeek V4 Pro 0423DeepSeek | Baidufp8 | $0.1977 / $0.3955 | $0.1775 / $0.3549 | −10.3%in −10.3% · out −10.3% |
| 2026-10-1707:37 UTC | DeepSeek V4 Pro 0423DeepSeek | Relacefp4 | $0.1979 / $0.4607 | $0.1675 / $0.4607 | −8.7%in −15.4% |
| 2026-10-1707:21 UTC | GLM 5.2Z.ai (Zhipu) | Relace | $0.0241 / $2.23 | $0.0131 / $2.23 | −1.4%in −45.6% |
| 2026-10-1707:05 UTC | Kimi K3Moonshot AI | Relacefp4 | $0.284 / $8.54 | $0.25 / $8.54 | −1.1%in −12.0% |
| 2026-10-1707:05 UTC | DeepSeek V4 Flash 0731DeepSeek | OpenInferencefp4 | $0.0189 / $0.9653 | $0.0154 / $0.9653 | −1.0%in −18.5% |
| 2026-10-1707:05 UTC | DeepSeek V4 Flash 0731DeepSeek | Relacefp4 | $0.018 / $0.9653 | $0.0154 / $0.9653 | −0.8%in −14.4% |
| 2026-10-1707:05 UTC | DeepSeek V4 Pro 0813DeepSeek | Baidufp8 | $0.2389 / $0.7168 | $0.2072 / $0.6217 | −13.3%in −13.3% · out −13.3% |
| 2026-10-1707:05 UTC | GLM 5.3Z.ai (Zhipu) | Relace | $0.13 / $2.20 | $0.069 / $2.20 | −7.1%in −46.9% |
| 2026-10-1706:49 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Relace | $0.0541 / $0.5578 | $0.0464 / $0.5578 | −3.2%in −14.2% |
| 2026-10-1706:49 UTC | GLM 5.3 FlashZ.ai (Zhipu) | OpenInferencefp4 | $0.057 / $0.5578 | $0.0464 / $0.5578 | −4.4%in −18.6% |
| 2026-10-1706:49 UTC | DeepSeek V4.1 FlashDeepSeek | Morphfp8 | $0.04 / $0.55 | $0.036 / $0.55 | −1.8%in −10.0% |
The 50 most recent of 523 cuts.
Price increases
| Date | Model | Provider and terms | Before | After | Change |
|---|---|---|---|---|---|
| 2026-10-1712:12 UTC | GLM 5.2Z.ai (Zhipu) | Relace | $0.0241 / $2.51 | $0.075 / $6.00 | +141.5%in +211.2% · out +139.5% |
| 2026-10-1712:12 UTC | DeepSeek V4 Flash 0731DeepSeek | Relacefp4 | $0.0071 / $0.2829 | $0.0058 / $1.28 | +326.5%in −18.3% · out +352.5% |
| 2026-10-1712:12 UTC | DeepSeek V4 Pro 0813DeepSeek | Relacefp4 | $0.1233 / $0.3831 | $0.12 / $8.00 | +1010.1%in −2.7% · out +1988.2% |
| 2026-10-1711:56 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Morphfp8 | $0.082 / $0.65 | $0.085 / $0.65 | +1.0%in +3.7% |
| 2026-10-1711:56 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Relace | $0.0238 / $0.3966 | $0.075 / $0.50 | +54.9%in +215.1% · out +26.1% |
| 2026-10-1711:56 UTC | GLM 5.3 FlashZ.ai (Zhipu) | OpenInferencefp4 | $0.0238 / $0.3966 | $0.075 / $0.50 | +54.9%in +215.1% · out +26.1% |
| 2026-10-1711:40 UTC | DeepSeek V4.1 FlashDeepSeek | Relace | $0.0153 / $0.594 | $0.20 / $0.60 | +87.5%in +1207.2% · out +1.0% |
| 2026-10-1711:40 UTC | DeepSeek V4.1 FlashDeepSeek | OpenInferencefp4 | $0.0153 / $0.594 | $0.20 / $0.60 | +87.5%in +1207.2% · out +1.0% |
| 2026-10-1711:24 UTC | DeepSeek V4 Pro 0423DeepSeek | Relacefp4 | $0.1675 / $0.4607 | $0.90 / $3.00 | +491.9%in +437.5% · out +551.2% |
| 2026-10-1711:24 UTC | DeepSeek V4 Flash 0423DeepSeek | OpenInferencefp4 | $0.0215 / $0.1388 | $0.0191 / $0.6992 | +272.1%in −11.2% · out +403.7% |
| 2026-10-1711:24 UTC | DeepSeek V4 Flash 0423DeepSeek | Relacefp4 | $0.0215 / $0.1388 | $0.0181 / $1.28 | +556.3%in −15.8% · out +822.2% |
| 2026-10-1711:08 UTC | GLM 5.2Z.ai (Zhipu) | Relace | $0.0241 / $2.37 | $0.0241 / $2.51 | +5.3%out +5.5% |
| 2026-10-1711:08 UTC | Kimi K3Moonshot AI | Morphfp8 | $0.30 / $14.90 | $0.80 / $14.90 | +9.5%in +166.7% |
| 2026-10-1710:52 UTC | GLM 5.3Z.ai (Zhipu) | Relace | $0.069 / $3.94 | $0.069 / $19.90 | +385.0%out +405.2% |
| 2026-10-1710:36 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Morphfp8 | $0.073 / $0.65 | $0.082 / $0.65 | +3.1%in +12.3% |
| 2026-10-1710:36 UTC | DeepSeek V4.1 FlashDeepSeek | Morphfp8 | $0.036 / $0.55 | $0.09 / $0.55 | +24.6%in +150.0% |
| 2026-10-1709:48 UTC | MiniMax H3 Max Text to VideoMiniMax | fal | $0.03 per second | $0.0325 per second | +8.3% |
| 2026-10-1709:48 UTC | H3 Max Turbo Text to VideoMiniMax | fal | $0.015 per second | $0.0163 per second | +8.3% |
| 2026-10-1709:48 UTC | H3 Max Turbo Image to VideoMiniMax | fal | $0.015 per second | $0.0163 per second | +8.3% |
| 2026-10-1709:48 UTC | H3 Max Image to VideoMiniMax | fal | $0.03 per second | $0.0325 per second | +8.3% |
| 2026-10-1709:48 UTC | H3 Max Camera ControlsMiniMax | fal | $0.03 per second | $0.0325 per second | +8.3% |
| 2026-10-1709:29 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Morphfp8 | $0.066 / $0.65 | $0.073 / $0.65 | +2.5%in +10.6% |
| 2026-10-1708:25 UTC | GLM 5.2Z.ai (Zhipu) | Relace | $0.0131 / $2.23 | $0.0241 / $2.37 | +7.6%in +84.0% · out +6.3% |
| 2026-10-1708:25 UTC | Kimi K3Moonshot AI | WaferUS regionmxfp4 | $0.36 / $14.90 | $0.50 / $14.90 | +2.6%in +38.9% |
| 2026-10-1708:25 UTC | Kimi K3Moonshot AI | Relacefp4 | $0.25 / $8.54 | $0.27 / $14.23 | +61.9%in +8.0% · out +66.7% |
| 2026-10-1708:09 UTC | GLM 5.3Z.ai (Zhipu) | Relace | $0.069 / $2.20 | $0.069 / $3.94 | +71.9%out +78.7% |
| 2026-10-1708:09 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Decartfp4 | $0.0825 / $0.275 | $0.129 / $0.43 | +56.4%in +56.4% · out +56.4% |
| 2026-10-1707:53 UTC | DeepSeek V4.1 FlashDeepSeek | Wafermxfp4 | $0.0299 / $0.65 | $0.10 / $0.65 | +28.4%in +234.4% |
| 2026-10-1707:53 UTC | DeepSeek V4.1 FlashDeepSeek | OpenInferencefp4 | $0.0309 / $0.4026 | $0.0256 / $0.6246 | +41.6%in −17.2% · out +55.1% |
| 2026-10-1707:53 UTC | DeepSeek V4.1 FlashDeepSeek | Relace | $0.0309 / $0.4026 | $0.0243 / $0.6246 | +40.8%in −21.4% · out +55.1% |
| 2026-10-1706:01 UTC | DeepSeek V4 Flash 0731DeepSeek | Inceptronfp4 | $0.13 / $0.4662 | $0.13 / $0.5978 | +15.4%out +28.2% |
| 2026-10-1706:01 UTC | DeepSeek V4 Flash 0731DeepSeek | OpenInferencefp4 | $0.07 / $0.14 | $0.0189 / $0.9653 | +192.0%in −73.0% · out +589.5% |
| 2026-10-1706:01 UTC | DeepSeek V4 Flash 0731DeepSeek | Relacefp4 | $0.07 / $0.14 | $0.018 / $0.9653 | +191.2%in −74.3% · out +589.5% |
| 2026-10-1705:45 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Morphfp8 | $0.055 / $0.65 | $0.069 / $0.65 | +5.2%in +25.5% |
| 2026-10-1705:45 UTC | GLM 5.3 FlashZ.ai (Zhipu) | OpenInferencefp4 | $0.075 / $0.50 | $0.057 / $0.5578 | +0.5%in −24.0% · out +11.6% |
| 2026-10-1704:57 UTC | DeepSeek V4 Flash 0423DeepSeek | OpenInferencefp4 | $0.0088 / $1.28 | $0.0215 / $1.28 | +2.9%in +144.3% |
| 2026-10-1704:25 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Inceptronfp8 | $0.10 / $0.69 | $0.10 / $0.8625 | +17.4%out +25.0% |
| 2026-10-1704:25 UTC | GLM 5.3 FlashZ.ai (Zhipu) | OpenInferencefp4 | $0.0392 / $0.30 | $0.075 / $0.50 | +73.6%in +91.3% · out +66.7% |
| 2026-10-1704:25 UTC | GLM 5.3 FlashZ.ai (Zhipu) | Morphfp8 | $0.049 / $0.65 | $0.055 / $0.65 | +2.3%in +12.2% |
| 2026-10-1704:09 UTC | DeepSeek V4.1 FlashDeepSeek | OpenInferencefp4 | $0.032 / $0.30 | $0.20 / $0.60 | +203.0%in +525.0% · out +100.0% |
| 2026-10-1703:37 UTC | DeepSeek V4 Flash 0423DeepSeek | OpenInferencefp4 | $0.0172 / $0.0363 | $0.0088 / $1.28 | +1386.4%in −48.8% · out +3427.3% |
| 2026-10-1703:21 UTC | DeepSeek V4 Flash 0731DeepSeek | OpenInferencefp4 | $0.00392 / $0.1436 | $0.07 / $0.14 | +125.3%in +1685.7% · out −2.5% |
| 2026-10-1703:21 UTC | GLM 5.3Z.ai (Zhipu) | Relace | $0.0608 / $4.40 | $0.75 / $2.40 | +1.5%in +1133.6% · out −45.5% |
| 2026-10-1703:05 UTC | GLM 5.3 FlashZ.ai (Zhipu) | OpenInferencefp4 | $0.0123 / $0.20 | $0.0392 / $0.30 | +76.2%in +218.2% · out +50.0% |
| 2026-10-1703:05 UTC | DeepSeek V4.1 FlashDeepSeek | OpenInferencefp4 | $0.032 / $0.144 | $0.032 / $0.30 | +65.0%out +108.3% |
| 2026-10-1702:33 UTC | DeepSeek V4 Pro 0423DeepSeek | Relacefp4 | $0.90 / $3.00 | $0.95 / $3.10 | +4.4%in +5.6% · out +3.3% |
| 2026-10-1702:33 UTC | DeepSeek V4 Flash 0423DeepSeek | Relacefp4 | $0.0163 / $1.28 | $0.0215 / $1.28 | +1.2%in +31.9% |
| 2026-10-1702:33 UTC | DeepSeek V4 Flash 0423DeepSeek | OpenInferencefp4 | $0.013 / $0.0363 | $0.0172 / $0.0363 | +16.5%in +31.9% |
| 2026-10-1702:17 UTC | DeepSeek V4 Flash 0731DeepSeek | Phala | $0.308 / $0.924 | $0.44 / $1.32 | +42.9%in +42.9% · out +42.9% |
| 2026-10-1702:01 UTC | DeepSeek V4 Pro 0813DeepSeek | Phala | $0.957 / $2.88 | $1.45 / $4.36 | +51.5%in +51.5% · out +51.5% |
The 50 most recent of 366 increases.
New priced listings
| Date | Model | Provider and terms | Price |
|---|---|---|---|
| 2026-10-1712:12 UTC | GLM 5.2Z.ai (Zhipu) | Nebius AI Studiofp4 | $1.40 / $4.40 |
| 2026-10-1712:12 UTC | Inkling SmallThinking Machines | DeepInfrafp8 | $0.45 / $1.20 |
| 2026-10-1712:12 UTC | DeepSeek V4 Flash 0731DeepSeek | OpenInferencefp4 | $0.0061 / $1.28 |
| 2026-10-1712:12 UTC | DeepSeek V4 Pro 0813DeepSeek | Alibaba | $1.12 / $3.37 |
| 2026-10-1712:12 UTC | DeepSeek V4 Pro 0813DeepSeek | DekaLLM | $0.88 / $3.80 |
| 2026-10-1712:12 UTC | Gemma 3 27BGoogle DeepMind | Nebius AI Studiofp8 | $0.10 / $0.30 |
| 2026-10-1711:56 UTC | GLM 5.3Z.ai (Zhipu) | Mistral AInvfp4 | $1.40 / $4.40 |
| 2026-10-1711:56 UTC | GLM 5.3 FlashZ.ai (Zhipu) | AtlasCloudfp8 | $0.15 / $0.50 |
| 2026-10-1711:56 UTC | Mistral Small 3.2 24BMistral AI | Venice AIfp8 | $0.0938 / $0.25 |
| 2026-10-1711:24 UTC | Qwen3.5 397B A17BQwen (Alibaba) | AtlasCloudfp8 | $0.55 / $3.50 |
| 2026-10-1711:24 UTC | Qwen3.5 397B A17BQwen (Alibaba) | Alibaba | $0.39 / $2.34 |
| 2026-10-1711:24 UTC | MiniMax M2.5MiniMax | Novita AIfp8 | $0.30 / $1.20 |
| 2026-10-1711:24 UTC | Qwen3.5-9BQwen (Alibaba) | Together AI | $0.17 / $0.25 |
| 2026-10-1711:24 UTC | Gemma 4 26B A4BGoogle DeepMind | GoogleGlobal endpoint | $0.15 / $0.60 |
| 2026-10-1711:24 UTC | Gemma 4 31BGoogle DeepMind | Novita AIbf16 | $0.14 / $0.40 |
| 2026-10-1711:24 UTC | MiMo-V2.5-ProXiaomi | StreamLake | $0.522 / $1.04 |
| 2026-10-1711:08 UTC | Gemini 3.5 FlashGoogle DeepMind | GoogleFlex tier · Global endpoint | $0.75 / $4.50 |
| 2026-10-1711:08 UTC | Nemotron 3.5 Super VLNVIDIA | Crusoebf16 | $0.10 / $0.40 |
| 2026-10-1711:08 UTC | DeepSeek V4.1 Flash 1017DeepSeek | InferenceNetfp8 | $0.10 / $0.50 |
| 2026-10-1711:08 UTC | Claude Sonnet 5Anthropic | AzureGlobal endpoint | $2.00 / $10.00 |
| 2026-10-1711:08 UTC | Kimi K3Moonshot AI | Together AI | $3.00 / $15.00 |
| 2026-10-1711:08 UTC | Gemini 3.5 Flash LiteGoogle DeepMind | GoogleFlex tier · Global endpoint | $0.15 / $1.25 |
| 2026-10-1710:52 UTC | DeepSeek V4 Flash 0731DeepSeek | Alibaba | $0.352 / $1.06 |
| 2026-10-1710:52 UTC | DeepSeek V4 Pro 0813DeepSeek | Together AI | $1.32 / $3.96 |
| 2026-10-1710:52 UTC | Muse Spark OpenMeta | Modal | $1.25 / $4.25 |
| 2026-10-1710:52 UTC | Llama 4 ScoutMeta | DeepInfraPriority tierfp8 | $0.15 / $0.45 |
| 2026-10-1710:52 UTC | GLM 5.3Z.ai (Zhipu) | ModelRunfp4 | $2.10 / $6.60 |
| 2026-10-1710:52 UTC | GLM 5.3Z.ai (Zhipu) | Nebius AI Studiofp4 | $1.40 / $4.40 |
| 2026-10-1710:52 UTC | Gemini 4 ArgonGoogle DeepMind | Google AI Studio | $2.00 / $10.00 |
| 2026-10-1710:36 UTC | GLM 5.3 FlashZ.ai (Zhipu) | CoreWeavenvfp4 | $0.15 / $0.50 |
| 2026-10-1710:36 UTC | Qwen3 235B A22B Instruct 2507Qwen (Alibaba) | Nebius AI Studiofp8 | $0.20 / $0.60 |
| 2026-10-1710:36 UTC | Gemini 3.8 FlashGoogle DeepMind | GoogleFlex tier · Global endpoint | $0.375 / $1.88 |
| 2026-10-1710:20 UTC | Qwen3 VL 30B A3B InstructQwen (Alibaba) | SiliconFlowfp8 | $0.29 / $1.00 |
| 2026-10-1710:20 UTC | DeepSeek V3.2DeepSeek | GMI Cloudfp8 | $0.2088 / $0.3096 |
| 2026-10-1710:20 UTC | GLM 4.7Z.ai (Zhipu) | Mancer 2fp4 | $0.70 / $2.50 |
| 2026-10-1710:20 UTC | GLM 4.7Z.ai (Zhipu) | DeepInfrafp4 | $0.40 / $1.75 |
| 2026-10-1710:20 UTC | Kimi K2.5Moonshot AI | Novita AI | $0.57 / $2.85 |
| 2026-10-1710:20 UTC | Claude Opus 4.6Anthropic | GoogleEurope region | $5.50 / $27.50 |
| 2026-10-1710:04 UTC | Grok 4.20 Multi-AgentxAI | xAI | $1.25 / $2.50 |
| 2026-10-1710:04 UTC | Gemma 4 26B A4BGoogle DeepMind | Makora | $0.10 / $0.18 |
| 2026-10-1710:04 UTC | GLM 5.1Z.ai (Zhipu) | Chutesfp8 | $0.98 / $3.08 |
| 2026-10-1710:04 UTC | DeepSeek V4 Pro 0423DeepSeek | Alibabafp8 | $1.42 / $2.83 |
| 2026-10-1709:49 UTC | MiniMax M3MiniMax | Venice AIfp8 | $0.30 / $1.20 |
| 2026-10-1709:49 UTC | Nova Micro 1.0Amazon | Amazon Bedrockeu-west-1 region | $0.035 / $0.14 |
| 2026-10-1709:49 UTC | GLM 5.2Z.ai (Zhipu) | Nebius AI Studiofp4 | $1.40 / $4.40 |
| 2026-10-1709:48 UTC | Kimi K3Moonshot AI | ParasailFast tierfp4 | $3.00 / $15.00 |
| 2026-10-1709:48 UTC | Llama 3.3 Euryale 70BSao10K | NextBitbf16 | $0.65 / $0.75 |
| 2026-10-1709:29 UTC | Llama 4 ScoutMeta | Novita AIbf16 | $0.18 / $0.59 |
| 2026-10-1709:29 UTC | Llama 4 ScoutMeta | DeepInfraFlex tierfp8 | $0.08 / $0.24 |
| 2026-10-1709:29 UTC | Qwen3.8 27BQwen (Alibaba) | Alibaba | $0.425 / $2.55 |
A provider endpoint priced for the first time. The 50 most recent of 287. New models are on New AI models.
Method
SF·SIX captures provider and aggregator prices continuously and records a change whenever a published input, output or native unit price of a priced offer differs from the previous capture (status, rule and metadata changes are not price changes). Token prices are USD per 1M tokens; “Before” and “After” show input / output. Moves under 0.5% are not listed, and OpenRouter’s automatic-routing endpoint is excluded because its price is a moving average over the providers it routes to. Only models and sources whose current data SF·SIX may publish are shown. Prices are USD per 1M tokens at the standard service tier in the default region: the cheapest qualifying provider per model, with flex, batch, priority and fast endpoints and unusually low outliers never used as the headline. “Blended” weights input and output 3:1.
Frequently asked questions
What was the biggest AI API price cut in the last 30 days?
DeepSeek V4.1 Flash on Baidu (fp8): $0.30 / $1.20 to $0.0357 / $0.1428 on 2026-10-16 (−88.1% at a 3:1 input:output mix).
How does SF·SIX detect AI API price changes?
Every capture of every provider price is compared with the previous one; a change in the published input, output or native unit price of a priced offer is recorded with its time. Moves under 0.5% and OpenRouter’s automatic-routing average (not a list price) are not listed. Changes are kept for 30 days.