MiniMax subscription

“Qwen, DS, GLM and MM are perfectly fine for REAL work”

“good on paper but fail to live up to expectations in practice”

“To this day Minimax and GLM are very strong coders but have the same issue, inference speed.”

“There are solid, cheap options like xAI, z.AI, MiniMax, Kimi, DeepSeek, and Alibaba.”

“Sitting around 15 minutes for local Minimax is stupid when you're trying to be productive.”

“I now make it a point of subscribing (directly) with providers that are transparent like MiniMax, DeepSeek, Xiaomi, & Z.ai.”

“MiniMax tends to think too much, but sometimes that makes a lot of sense, too.”

“there's many good Chinese models and labs, glm, qwen, deepseek, kimik, mimo, longcat, minimax”

“MiniMax and MiMo are a lot more reliable (and cheap)”

“After 75% price cuts by DeepSeek & Xiaomi MiMo (and now MiniMax), I pretty much packed my Claude bag up and moved over.”

“budget alternatives from the families of DeepSeek, Kimi, GPT, MiniMax, etc. which are very much Sonnet grade”

“MiMo and DeepSeek, even MiniMax, are way cheaper and arguably better”

“much, much less prone to 40k tokens of monologue before generating an answer”

“MiniMax is kind of horrible that it somewhat regularly fall into loops”

“MiniMax got stuck with looping through broken tool calls all the time”

“There are larger open weights models that are more capable, such as minimax and glm5.1.”

“MiniMax develops one of world's leading open source LLMs.”

“you can plug deepseek/kimi/minimax into claude code just fine”

“later GLM, Kimi and Minimax succeeded on without issues”

“arguably MiniMax are the ones that are perhaps falling behind”

“Kimi, Mimimax and GLM models provide far more robust coding assistance at sometimes no cost”

“MiniMAX and Claude models tend to be better AI models at doing agentic tasks.”

“Many of them (K2.5 the Minimax, the GLM models, the Qwen 3.6 models) are about as capable as frontier US models”

“In my harness Minimax wasn't even getting half of them right, whereas Sonnet was 100%.”

“Together with GLM, Qwen and Minimax they are an amazing backstop just the way they are right now”

“they've all been barely half as capable as a middling Codex model”

“Minimax generating a single output stream at its max throughput 24/7 only comes to about $90/mo.”

“for anything even remotely complex these models seriously struggle.”

“I didn't notice slower outputs, it's even faster than Anthropic.”

“The down sides emerge pretty fast: much higher reasoning token use, slower outputs”

“will allow you to mess around with this stuff for negligible cost”

“MiniMax, GLM and Kimi K2 seem to be better models for this purpose than Qwen”

“MiniMax, DeepSeek, and Moonshot are all releasing models for the public to use for free.”

“I can get Minimax, Kimi, and Glm for the fraction of the price”

“Gemini/MiniMax are better value models that still perform well”

“I ranked Minimax higher for its agentic capabilities”

“The current SotA for open models (dsv3, glm, minimax, devstral, etc) are at or above the mini versions of top labs”

“you can download them and they're still very capable”

“Minimax has been great for super high speed web/js/ts related work.”

“Kimi, GLM, QWen and MiniMax are all frontier models no matter how you judge it.”