MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second

Re:MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per s... by SwellJoe on Monday June 08, 2026 @03:18PM (#48450197)

DeepSeek is the fastest model in the benchmarks I've been doing (https://swelljoe.com/post/will-it-mythos/). Followed not so closely by Opus 4.8 and even less closely by Gemini 3.5 Flash and GPT 5.5. I've been really impressed with it, so far. It's also among the best at doing the work, though still trailing the frontier models from Anthropic and OpenAI.

[ Parent | on HN ]

Re:MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per s... by anschl on Tuesday June 09, 2026 @04:20AM (#48458224)

Nice benchmark, thanks! Which quants did you choose for the self hosted models?

[ Parent | on HN ]

Re:MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per s... by SwellJoe on Tuesday June 09, 2026 @04:50AM (#48458442)

8-bit on that one (unsloth 8_K_XL). But, the next post compares all common quantizations of Qwen 3.6.

I have another coming in a day or so for Gemma 4 with the 4-bit QAT version, which is very surprising (in a good way, Gemma 4 is impressive for this task).

[ Parent | on HN ]