MiniMaxAI/MiniMax-M2.5 · Discussions

Production deployment at scale

#55 opened about 3 hours ago by

O96a

where is minimax 2.7

1

#54 opened 3 days ago by

devops724

minimax2.5

#52 opened 5 days ago by

gsting

关于SWE-Bench-Pro复现的疑问

1

#51 opened 9 days ago by

Clumsy1

Add SWE-Bench Pro evaluation result

4

#50 opened 10 days ago by

nielsr

can we get minimax-m2.7

🤗 12

5

#49 opened 10 days ago by

CHNtentes

[Model bug] Typos issues with either quantization or Fp8 with transformers v5 (or both).

👍 3

2

#48 opened 11 days ago by

mratsim

Add MiniMax reported SWE-Bench Verified result

#47 opened 12 days ago by

nielsr

Does anyone know a good 4 bit quant version?

#46 opened 14 days ago by

pathosethoslogos

Install & run MiniMaxAI/MiniMax-M2.5 easily using llmpm

👍 1

#45 opened 17 days ago by

sarthak-saxena

Security/Compliance Audit: EU AI Act & NIST Exposure

1

#44 opened 20 days ago by

tradeapollo

BFCL results

#41 opened 28 days ago by

paverGulyaevich

Transformers v5 support

#40 opened 29 days ago by

AntonV

Could I deploy the model with 8XA100？

#37 opened about 1 month ago by

Yuxin362

what lisencing does the MiniMaxAI/MiniMax-M2.5 follows?

👍 1

1

#36 opened about 1 month ago by

naren-1219

OctoBench performance

#35 opened about 1 month ago by

petrovalex

Which reasoning parser to use?

👍 3

1

#33 opened about 1 month ago by

g-a-b-y

more about this model

2

#27 opened about 1 month ago by

surabb

What hardware needed for 1$ per hour?

🔥 1

#26 opened about 1 month ago by

kabachuha

perf degradation between vllm 0.14.0 to 0.15.1 (and nightly)

4

#25 opened about 1 month ago by

Meital

Learning

#24 opened about 1 month ago by

Diallex

This model start making repeated mistakes affter 90,000 tokens on opencode

#23 opened about 1 month ago by

acharyaaditya26

salam

#21 opened about 1 month ago by

Nimolam

关于GPQA benchmark的一些疑问

1

#18 opened about 1 month ago by

zymu

关于 swebench 复现的疑问

4

#17 opened about 1 month ago by

hrw

Can we have a smaller version version than the 230B / 10B active MoE e.g 120B / 10B active

👍 6

2

#15 opened about 1 month ago by

KingBadger

Recommended max_kv_size for MLX deployment to prevent OOM on Apple Silicon

👍 2

1

#14 opened about 1 month ago by

Neo2025new

VIBE Pro benchmark release date?

👍 1

1

#13 opened about 1 month ago by

ZeroAurora

This version is the gateway to AGI of open weights community

🚀 4

#11 opened about 1 month ago by

rosspanda0

Just twiddling my thumbs

➕ 2

3

#10 opened about 1 month ago by

huggyfaceenjoyer

GGUFs working well on ik/llama.cpp!

🚀 7

5

#9 opened about 1 month ago by

ubergarm

Inconsistent shard naming (124 vs 126) in both files and index.json

👀 4

1

#8 opened about 1 month ago by

cmadh

MiniMax-M2.5 MLX q3/q4 uploaded

🚀 1

1

#7 opened about 1 month ago by

ahoybrotherbear

Thank you!

❤️ 13

1

#6 opened about 1 month ago by

CHONGYOEYAT

context window

2

#5 opened about 1 month ago by

GustavB7

mtp

➕ 10

3

#4 opened about 1 month ago by

CHNtentes

Will M2.5-Lightning be released?

👍 4

3

#2 opened about 1 month ago by

Exostyle81