Minimax GGUF models

251 GGUF models in the Minimax family, quantized by 132 uploaders. Sizes run from 20M to 426B parameters. The longest trained context is 1024K tokens.

TinySmallMediumLarge
Most downloadedNewestLargestLongest contextSmallest
MI
MiniMax H3
Minimax · unsloth
Params20B
Context—
Q4_K_M18.2 GB
GGUF
MI
MiniMax H3
Minimax · DeepBeepMeep
Params25.8B
Context—
Q4_K_M14.6 GB
GGUF
MI
MiniMax-Music3 flow matching DiT
Minimax · Serveurperso
Params2.4B
Context—
Q4_K_M1.4 GB
GGUF
MI
MiniMax H3 Pruned
Minimax · Abiray
Params20.1B
Context—
Q4_K_M11.6 GB
GGUF
MI
MiniMax-Music
Minimax · audio-cpp
Params4.9B
Context—
Smallest0.1 GB
GGUF
MI
MiniMax H3 GGUFs
Minimax · realrebelai
Params33.1B
Context—
Q4_K_M14.6 GB
GGUF
MI
MiniMax H3
Minimax · leejet
Params20.1B
Context—
Q4_K_M11.4 GB
GGUF
MI
MiniMax H3
Minimax · Abiray
Params25.8B
Context—
Q4_K_M14.6 GB
GGUF
MI
MiniMax H3
Minimax · molbal
Params20.1B
Context—
Smallest11.4 GB
GGUF
MI
Sakura MiniMax H3 Uncensored
Minimax · webmp3
Params25.8B
Context—
Q4_K_M15.1 GB
GGUF
MI
MiniMax H3 encoder
Minimax · joeygambino
Params32.8B
Context256K
Q4_K_M19.8 GB
CodingVision
MI
MiniMax H3 curve
Minimax · joeygambino
Params20.1B
Context—
Smallest11.5 GB
GGUF
MI
minimax_h3_turbo_f16
Minimax · ChrisColeTech
Params20.1B
Context—
Q4_K_M11.4 GB
GGUF
MI
Sakura MiniMax H3
Minimax · webmp3
Params20.1B
Context—
Q4_K_M12.6 GB
GGUF
MI
MiniMax M3.1 Flash Maya
Minimax · peasantsmith
Params426B
Context1024K
Smallest93.0 GB
ReasoningTool use
MI
MiniMax-M3.1-Flash
Minimax · LyJonathan
Params426B
Context1024K
Smallest94.8 GB
ReasoningTool useVision
MI
MiniMax M3.1 Flash
Minimax · unsloth
Params426B
Context1024K
Smallest0.5 GB
GGUF
MI
MiniMax H3 Singularity
Minimax · Abiray
Params20.1B
Context—
Q4_K_M11.6 GB
GGUF
MI
Minimax-M3
Minimax · unsloth
Params426B
Context1024K
Q4_K_M264.9 GB
Vision
MI
minimax_music3_dit_Q6_K
Minimax · ChrisColeTech
Params7.61B
Context—
Smallest2.1 GB
GGUF
MI
MiniMax H3 comfyUI
Minimax · vantagewithai
Params33.1B
Context—
Q4_K_M19.9 GB
GGUF
MI
Minimax-M2
Minimax · unsloth
Params230B
Context192K
Q4_K_M138.3 GB
GGUF
MI
MiniMax H3 Pruned Ref Delta Fused
Minimax · Abiray
Params20.1B
Context—
Q4_K_M11.6 GB
GGUF
MI
OpenYourMind_Minimax M3 Abliterated Clean
Minimax · Sica-Quants-2
Params426B
Context1024K
Smallest155.0 GB
ReasoningVision
MI
MiniMax M2.7
Minimax · bartowski
Params230B
Context192K
Q4_K_M138.8 GB
GGUF
MI
FastVideo FastH3 Omni 8 Step V1 ComfyUI
Minimax · vantagewithai
Params22.1B
Context—
Q4_K_M13.7 GB
GGUF
MI
Minimax-M2.7
Minimax · unsloth
Params230B
Context192K
Q4_K_M140.2 GB
GGUF
MI
MiniMax-H3 fl2va
Minimax · joeygambino
Params33.1B
Context—
Smallest19.9 GB
GGUF
MI
Minimax-M2.1
Minimax · unsloth
Params230B
Context192K
Q4_K_M138.3 GB
GGUF
MI
MiniMax H3 Turbo
Minimax · molbal
Params20.1B
Context—
Smallest11.4 GB
GGUF

About the Minimax GGUF models

Minimax is represented on ggufmodels.com by 251 GGUF models, quantized by 132 uploaders on Hugging Face. The most downloaded are MiniMax H3, MiniMax H3 and MiniMax-Music3 flow matching DiT. Parameter counts run from 20M to 426B, and the longest trained context window is 1024K tokens.

A Q4_K_M file of a Minimax model is between 0.0 to 283.6 GB, which is the quant most people start with. Most files come from mradermacher (29), DevQuasar (9) and bartowski (8). The quants you will meet most often are Q4_K_M (133), Q8_0 (118), Q6_K (102) and Q5_K_M (91), where the number is how many models offer that quant.

Frequently asked questions

Which Minimax GGUF should I download?

The most downloaded Minimax GGUF here is MiniMax H3. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.

How much VRAM does a Minimax model need?

Q4_K_M files of Minimax models on this site range from 0.0 to 283.6 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.

Where do Minimax GGUF files come from?

The most common quantizers here are mradermacher, DevQuasar and bartowski. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.

Keep browsing