← Ministral 3

Ministral 3 14B-INSTRUCT-2512

No machine on this network has measured this model yet. Everything below is what the catalog knows exists — where the weights come from, how large each build is, and whose work it is. Nothing here is a claim about what the model can do; that only appears once a machine has run it and signed the result.

mlx-community/mistralai_Ministral-3-14B-Instruct-2512-MLX-MXFP4mlx · 4BIT · 6.7 GiB · quantizer (mlx-community)
hf.co/mistralai/Ministral-3-14B-Instruct-2512-GGUF:Q4_K_Mollama · Q4_K_M · 7.7 GiB · vendor (mistralai)
hf.co/mistralai/Ministral-3-14B-Instruct-2512-GGUF:Q5_K_Mollama · Q5_K_M · 9.0 GiB · vendor (mistralai)
hf.co/mistralai/Ministral-3-14B-Instruct-2512-GGUF:Q8_0ollama · Q8_0 · 13.4 GiB · vendor (mistralai)
cyankiwi/Ministral-3-14B-Instruct-2512-AWQ-4bitvllm · W4A16 · 19.4 GiB · quantizer (cyankiwi)
hf.co/mistralai/Ministral-3-14B-Instruct-2512-GGUF:BF16ollama · BF16 · 25.2 GiB · vendor (mistralai)
RedHatAI/Ministral-3-14B-Instruct-2512-FP8-dynamicvllm · FP8 · 31.3 GiB · quantizer (RedHatAI)
mistralai/Ministral-3-14B-Instruct-2512-BF16vllm · BF16 · 51.9 GiB · vendor (mistralai)
derivatives — somebody else's change to these weights

Fine-tunes, merges and abliterations of this member, each its own model with its own page. None of their numbers are counted above.

Ministral-3-14B-Instruct-2512-bnb-4bitnot measured