Baseten

Inference platform for deploying and serving AI models

baseten.coSan Francisco, United States

Follow
Baseten website cover
Baseten is a San Francisco company, founded in 2019, that provides infrastructure for deploying and serving machine learning models in production. It also publishes a small number of in-house models on its platform.

Catalogue

Available models

Headquarters: US · 18 models · 31 endpoints

ModelEndpointContextInput /MOutput /MCached input /MCache write /MUptime 1d
Beam 1.0 501B A23Bbaseten/fp41M$0.40$1.20$0.04—97.75%
DeepSeek V4 Flash 0731baseten/fp321M$0.13$0.26$0.028—98.15%
DeepSeek V4 Flash 0731baseten/fp81M$0.13$0.26$0.028—98.39%
DeepSeek V4 Pro 0813baseten/fp41M$1.32$3.96$0.132—98.51%
DeepSeek V4.1 Flashbaseten1M$0.30$1.20$0.007—97.29%
DeepSeek V4.1 Flashbaseten/us1M$0.30$1.20$0.007—94.55%
DeepSeek V4.1 Flash 1017baseten1M$0.30$1.20$0.007—98.54%
DeepSeek V4.1 Flash 1017baseten/fast1M$0.60$2.40$0.014—98.19%
Elephant 1baseten16K$0.035$0.00——98.44%
GLM 5.2baseten/fast1M$2.10$6.60$0.21—97.67%
GLM 5.2baseten/fp81M$1.40$4.40$0.14—95.48%
GLM 5.2baseten/fp8/us1M$1.40$4.40$0.14—98.87%
GLM 5.2baseten/us/fast1M$2.10$6.60$0.21—97.70%
GLM 5.3baseten/fast1M$2.10$6.60$0.21—98.92%
GLM 5.3baseten/fp41M$1.40$4.40$0.14—98.54%
GLM 5.3baseten/fp4/us1M$1.40$4.40$0.14—98.27%
GLM 5.3baseten/us/fast1M$2.10$6.60$0.21—97.87%
GLM 5.3 Flashbaseten/fp81M$0.15$0.50$0.03—98.47%
GLM 5.3 Flashbaseten/fp8/us1M$0.15$0.50$0.03—0.00%
Joey 1baseten16K$0.01$0.00——97.87%
Kimi K3baseten/fp81M$3.00$15.00$0.30—97.00%
Mammoth 1baseten16K$0.10$0.00——98.89%
Muse Spark Openbaseten/fp41M$1.25$4.25$0.15—98.22%
Nemotron 3 Ultrabaseten/fp4203K$0.60$2.40$0.12—96.92%
Nemotron 3 Ultrabaseten/fp4/us203K$0.60$2.40$0.12—97.65%
Nemotron 3.5 Super VLbaseten/fp41M$0.20$0.50$0.04—95.90%
gpt-oss-120bbaseten/fp4128K$0.10$0.50$0.10—97.07%
gpt-oss-120bbaseten/fp4/us128K$0.10$0.50$0.10—98.47%

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.