Back/Baseten

Baseten

San Francisco-based AI infrastructure company founded in 2019. Private. Operates in AI model inference and deployment. Serves developers and enterprises with GPU-backed model serving, dedicated deployments, and training infrastructure.

Plans tracked

no add-ons

Pricing changes

Latest Wk 40 '26

Pricing model

Usage-based

No free tier

Pricing Change History

3 changes

20263 changes

Wk 40 '26

Packaging

Baseten adds GLM-5.3 Fast, retires 6 models, cuts DeepSeek V4.1 Flash cache price 77%

New model GLM-5.3 Fast added to the Model APIs price list at $2.10 input / $0.21 cache input / $6.60 output per 1M tokens. DeepSeek V4.1 Flash cache-input price cut from $0.03 to $0.007 per 1M tokens (-77%); input $0.30 and output $1.20 unchanged.

View diff

Wk 32 '26

Improvement

Baseten: added Kimi K3, Inkling-Small & DeepSeek-V4-Flash-0731 model APIs

Added 3 new models to Model APIs: Kimi K3 ($3.00/$0.30/$15.00 per 1M input/cache/output tokens), Inkling-Small ($0.50/$0.10/$1.20), DeepSeek-V4-Flash-0731 ($0.13/$0.028/$0.26)

View diff

Wk 31 '26

Price decrease

Baseten: GLM 5.2 cache-input price cut from $0.26 to $0.14/1M tokens

GLM 5.2 cache input token pricing decreased from $0.26 to $0.14 per 1M tokens (input $1.40 and output $4.40 unchanged)

View diff