AxForge
Sign in Get started

Model reference · open weights

MobileMoE-S-SFT

LLMs facebook Text gen · MoE 1 build Non-commercial 13 dl/mo

MobileMoE-S-SFT is an open-weight language model from Meta. MobileMoE-S-SFT (BF16) weighs 2.5 GB; the smallest configuration that runs it is RTX 3060 12 GB.

  • MobileMoE-S-SFT is an instruction-tuned on-device Mixture-of-Experts language model developed by Meta for text generation.
  • It features 272 million active parameters out of 1.3 billion total, supports a context length of 8,192 tokens, and is trained for English.
  • The model is distributed under the FAIR Noncommercial Research License.

Summary of the facebook/MobileMoE-S-SFT model card, 2026-10-01

Hugging Face ↗Paper ↗

What it is

Released byMeta
Released2026-06-11
Parameters1.3B
VRAM2.5 GB for the weights

What it runs on

Memory and cards for MobileMoE-S-SFT (BF16)

2.5 GBweights, file size
762 MBruntime overhead, at least

How much memory each request adds isn't estimated yet for this architecture. The weights need at least the cards below, plus room for the context.

CardWeights alone
RTX 3060 12 GBfits
RTX 4060 Ti 16 GBfits
RTX 3090 24 GBfits
RTX 4090 24 GBfits
RTX 5090 32 GBfits
L40S 48 GBfits
A100 80 GBfits
H100 80 GBfits
RTX PRO 6000 Blackwell 96 GBfits
DGX Spark (GB10) 128 GB unifiedfits
H200 141 GBfits
B200 180 GBfits

How it works

How language models work

Your prompttext / messagesTransformerattention over tokensNext-token loopgenerate + streamResponsetext · tool callsA language model reads your tokens and predicts the next one, again and again, streaming the reply back.
© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms