AaryanK / minimax family / text-generation

minimax-m2.7

minimax/minimax-m2.7

GGUF Provisional scores Updated 12 Apr 2026

Run locally Smallest indexed GGUF / needs Ollama

ollama run hf.co/minimax/minimax-m2.7/MiniMaxAI_MiniMax-M2.7-IQ1_S.gguf

Benchmarks

Where it scores.

Methodology →

General Provisional

89.4

Coding Provisional

89.4

Math Provisional

89.4

Reasoning Provisional

89.4

Normalized 0–100 across indexed models.

Runnable variants

Pick a size.

26 indexed

Precision	Format	Size	Params	File
IQ1_S / smallest	GGUF	46.7 GB	—	MiniMaxAI_MiniMax-M2.7-IQ1_S.gguf
IQ1_M	GGUF	52.2 GB	—	MiniMaxAI_MiniMax-M2.7-IQ1_M/MiniMaxAI_MiniMax-M2.7-IQ1_M-00001-of-00002.gguf
IQ2_XXS	GGUF	60.8 GB	—	MiniMaxAI_MiniMax-M2.7-IQ2_XXS/MiniMaxAI_MiniMax-M2.7-IQ2_XXS-00001-of-00002.gguf
IQ2_XS	GGUF	67.9 GB	—	MiniMaxAI_MiniMax-M2.7-IQ2_XS/MiniMaxAI_MiniMax-M2.7-IQ2_XS-00001-of-00002.gguf
IQ2_S	GGUF	69.1 GB	—	MiniMaxAI_MiniMax-M2.7-IQ2_S/MiniMaxAI_MiniMax-M2.7-IQ2_S-00001-of-00002.gguf
IQ2_M	GGUF	76.4 GB	—	MiniMaxAI_MiniMax-M2.7-IQ2_M/MiniMaxAI_MiniMax-M2.7-IQ2_M-00001-of-00002.gguf
Q2_K	GGUF	80.1 GB	—	MiniMaxAI_MiniMax-M2.7-Q2_K/MiniMaxAI_MiniMax-M2.7-Q2_K-00001-of-00003.gguf
Q2_K_L	GGUF	80.7 GB	—	MiniMaxAI_MiniMax-M2.7-Q2_K_L/MiniMaxAI_MiniMax-M2.7-Q2_K_L-00001-of-00003.gguf
IQ3_XXS	GGUF	95.2 GB	—	MiniMaxAI_MiniMax-M2.7-IQ3_XXS/MiniMaxAI_MiniMax-M2.7-IQ3_XXS-00001-of-00003.gguf
Q3_K_S	GGUF	99.3 GB	—	MiniMaxAI_MiniMax-M2.7-Q3_K_S/MiniMaxAI_MiniMax-M2.7-Q3_K_S-00001-of-00003.gguf
IQ3_XS	GGUF	104.1 GB	—	MiniMaxAI_MiniMax-M2.7-IQ3_XS/MiniMaxAI_MiniMax-M2.7-IQ3_XS-00001-of-00003.gguf
Q3_K_M	GGUF	104.1 GB	—	MiniMaxAI_MiniMax-M2.7-Q3_K_M/MiniMaxAI_MiniMax-M2.7-Q3_K_M-00001-of-00003.gguf
Q3_K_L	GGUF	108.5 GB	—	MiniMaxAI_MiniMax-M2.7-Q3_K_L/MiniMaxAI_MiniMax-M2.7-Q3_K_L-00001-of-00003.gguf
IQ3_M	GGUF	108.8 GB	—	MiniMaxAI_MiniMax-M2.7-IQ3_M/MiniMaxAI_MiniMax-M2.7-IQ3_M-00001-of-00003.gguf
Q3_K_XL	GGUF	109.0 GB	—	MiniMaxAI_MiniMax-M2.7-Q3_K_XL/MiniMaxAI_MiniMax-M2.7-Q3_K_XL-00001-of-00003.gguf
IQ4_XS	GGUF	122.4 GB	—	MiniMaxAI_MiniMax-M2.7-IQ4_XS/MiniMaxAI_MiniMax-M2.7-IQ4_XS-00001-of-00004.gguf
IQ4_NL	GGUF	129.5 GB	—	MiniMaxAI_MiniMax-M2.7-IQ4_NL/MiniMaxAI_MiniMax-M2.7-IQ4_NL-00001-of-00004.gguf
Q4_0	GGUF	129.8 GB	—	MiniMaxAI_MiniMax-M2.7-Q4_0/MiniMaxAI_MiniMax-M2.7-Q4_0-00001-of-00004.gguf
Q4_K_S	GGUF	133.8 GB	—	MiniMaxAI_MiniMax-M2.7-Q4_K_S/MiniMaxAI_MiniMax-M2.7-Q4_K_S-00001-of-00004.gguf
Q4_K_M	GGUF	138.8 GB	—	MiniMaxAI_MiniMax-M2.7-Q4_K_M/MiniMaxAI_MiniMax-M2.7-Q4_K_M-00001-of-00004.gguf
Q4_K_L	GGUF	139.3 GB	—	MiniMaxAI_MiniMax-M2.7-Q4_K_L/MiniMaxAI_MiniMax-M2.7-Q4_K_L-00001-of-00004.gguf
Q4_1	GGUF	143.5 GB	—	MiniMaxAI_MiniMax-M2.7-Q4_1/MiniMaxAI_MiniMax-M2.7-Q4_1-00001-of-00004.gguf
Q5_K_S	GGUF	157.7 GB	—	MiniMaxAI_MiniMax-M2.7-Q5_K_S/MiniMaxAI_MiniMax-M2.7-Q5_K_S-00001-of-00004.gguf
Q5_K_M	GGUF	162.7 GB	—	MiniMaxAI_MiniMax-M2.7-Q5_K_M/MiniMaxAI_MiniMax-M2.7-Q5_K_M-00001-of-00005.gguf
Q6_K	GGUF	197.1 GB	—	MiniMaxAI_MiniMax-M2.7-Q6_K/MiniMaxAI_MiniMax-M2.7-Q6_K-00001-of-00005.gguf
Q8_0	GGUF	243.1 GB	—	MiniMaxAI_MiniMax-M2.7-Q8_0/MiniMaxAI_MiniMax-M2.7-Q8_0-00001-of-00007.gguf

Sizes are file sizes, not peak runtime memory.

Raw benchmarks

The inputs.

1 records

Source	Task	Raw	Normalized
Arena	ArenaScore	1415.0	89.4

From the Hugging Face model card

About this model.

Read it on Hugging Face ↗