Model index / Other / Flint-40B-A2.5B-abliterated
Other / MODEL PROFILE

Flint-40B-A2.5B-abliterated

dima6312 · Published release · Abliteration · GGUF / MLX / EXL3 / Transformers / FP8

Model familyOther
Parameters40B
Estimated Q4 VRAMAbout 24 GB
30-day downloads20.4K
Declared licenseapache-2.0
Source repositories9

Model source

Published by dima6312 on Hugging Face. This index organizes public repository metadata for discovery and comparison. Refer to the publisher and independent testing for actual model behavior.

Declared base model

RecursiveSI/flint ↗

Hardware and context

Estimated at 24 GB VRAM for 40B parameters with 4-bit weights, including 15% weight overhead and 1 GB of baseline headroom. KV cache, context length, multimodal components and concurrency increase actual memory usage.The parameter count was parsed from the model name and has not been independently verified.

Context length: Not stated in public metadata. Please consult the model card.

Download sources and quantized variants

9 repositories
Uncensored is a publisher description, not a guarantee of zero refusals, capability, accuracy or safety. This index does not measure refusal rates or performance benchmarks.

More from this family

More Other →