MPT-7B

MosaicML · 2023 · LLM · USA

A to Z

A · App: No
B · Built By: MosaicML
C · CEO & Politics: Ali Ghodsi — No public partisan activity documented
D · Date: 2023
E · Expertise: Intermediate+
F · Free Tier: Yes
G · Generates: Text only
H · Deployment: Cloud + Local (open weights)
I · Inputs: Text
J · Jurisdiction: USA
K · Key Staffers: Ali Ghodsi (CEO), Matei Zaharia (Co-founder), Jonathan Frankle (Chief Scientist, Mosaic AI)
L · Languages: English-primary
M · Model Type: LLM
N · Safety Issues: None reported
O · Open Source: Yes — open weights
P · Pricing: Retired (no longer sold)
Q · AIDb Rating: [UNKNOWN]
R · What Makes It Special: MPT-7B, the first model in MosaicML's Foundation Series: a 6.7B-parameter decoder-style transformer trained from scratch on 1 trillion tokens of text and code in 9.5 days on the MosaicML platform, with no human intervention, for about $200,000. Open source under Apache 2.0 and licensed for commercial use, it matched LLaMA-7B in MosaicML's evaluations and uses FlashAttention and ALiBi; the MPT-7B-StoryWriter-65k+ variant handles 65K-token contexts, and Instruct and Chat versions shipped too.
S · Stock / Status: Private
T · Training Cutoff: [UNKNOWN]
U · Users: [UNKNOWN]
V · Voice Chat: No
W · Weights: Downloadable
X · X Handle: @databricks
Y · Year Founded: [UNKNOWN]
Z · Bottom Line: Open-weights LLM from MosaicML, released 2023.

Data completeness 85% · Last verified 2026-05-12 · Primary source ↗

“MPT-7B.” AIDb — The AI Database. Verified 2026-05-12. https://theaidatabase.com/m/mpt-7b.html