Icon for FlexAI

FlexAI

Open-weight models through an OpenAI-compatible API, with EU-only workspaces and dedicated GPUs

FlexAI serves open-weight models through Token Factory, an OpenAI-compatible API priced per model per million tokens. The catalogue covers text, code, vision, embedding, speech and image models, including gpt-oss, Llama, DeepSeek, Qwen, GLM, Gemma and MiniMax.

Inference runs on NVIDIA, AMD and Tenstorrent hardware across owned and partner capacity, with zero retention of prompts and outputs. EU workspaces are billed in euros by FlexAI SAS in Paris and served only from EU infrastructure. Dedicated GPUs are also available by the minute or reserved monthly.

Pay as you go, from $0.018/M input tokens on GPT-OSS 20B.

Pricing: Per token usage

Pricing $0.018/M input tokens
HQ 🇺🇸 United States
Compliance SOC 2 · GDPR

Data and privacy

Checked 19 November 2026
Data processed
EU Workspaces: processed in the EU via eu.api.flex.ai, never routed to non-EU inference; shared account and admin services may run globally source
Sub-processors
Not public
US parent company
Yes

Read from FlexAI's own legal pages. "Not public" means it is not on their site; some providers share it on request.

Screenshot of FlexAI webpage

FlexAI prices by model

Per 1M tokens, read off FlexAI's own pricing page on the date shown.

Model Input / 1M Cached / 1M Output / 1M Checked Notes
DeepSeek V4.1 Flash 1017 $0.08 $0.0055 $0.605 19 Nov 2026
GLM 5.3 Flash $0.088 $0.0132 $0.25 19 Nov 2026
Gemma 4 26B A4B $0.067 $0.01 $0.29 19 Nov 2026
Gemma 4 31B $0.14 $0.021 $0.40 19 Nov 2026
Llama 3.3 70B Instruct $0.22 $0.033 $0.50 19 Nov 2026
MiniMax M3.1 Flash $0.15 $0.015 $0.60 19 Nov 2026
Qwen3.8 27B $0.20 $0.03 $2.50 19 Nov 2026
gpt-oss-120b $0.03 $0.0045 $0.15 19 Nov 2026
gpt-oss-20b $0.018 $0.0027 $0.09 19 Nov 2026

Work on FlexAI? Feature it at the top of Inference APIs.

Is your product missing?

Add it here →