TeichAI / mimo family
mimo-v2-flash
xiaomi/mimo-v2-flash
Run locally Smallest indexed GGUF / needs Ollama
ollama run hf.co/xiaomi/mimo-v2-flash/MiMo-V2-Flash-UD-TQ1_0.gguf Benchmarks
Where it scores.
General Provisional
70.6
Coding Provisional
65.4
Math Provisional
68.3
Reasoning Provisional
52.5
Normalized 0–100 across indexed models.
Runnable variants
Pick a size.
| Precision | Format | Size | Params | File |
|---|---|---|---|---|
| Q4 / smallest | GGUF | 77.7 GB | — | MiMo-V2-Flash-UD-TQ1_0.gguf |
| IQ1_S | GGUF | 87.2 GB | — | UD-IQ1_S/MiMo-V2-Flash-UD-IQ1_S-00001-of-00002.gguf |
| IQ1_M | GGUF | 93.8 GB | — | UD-IQ1_M/MiMo-V2-Flash-UD-IQ1_M-00001-of-00002.gguf |
| IQ2_XXS | GGUF | 101.6 GB | — | UD-IQ2_XXS/MiMo-V2-Flash-UD-IQ2_XXS-00001-of-00003.gguf |
| IQ2_M | GGUF | 106.7 GB | — | UD-IQ2_M/MiMo-V2-Flash-UD-IQ2_M-00001-of-00003.gguf |
| Q2_K | GGUF | 112.3 GB | — | Q2_K/MiMo-V2-Flash-Q2_K-00001-of-00003.gguf |
| Q2_K_L | GGUF | 112.4 GB | — | Q2_K_L/MiMo-V2-Flash-Q2_K_L-00001-of-00003.gguf |
| Q2_K_XL | GGUF | 116.3 GB | — | UD-Q2_K_XL/MiMo-V2-Flash-UD-Q2_K_XL-00001-of-00003.gguf |
| IQ3_XXS | GGUF | 127.5 GB | — | UD-IQ3_XXS/MiMo-V2-Flash-UD-IQ3_XXS-00001-of-00003.gguf |
| Q3_K_S | GGUF | 133.1 GB | — | Q3_K_S/MiMo-V2-Flash-Q3_K_S-00001-of-00003.gguf |
| Q3_K_XL | GGUF | 137.1 GB | — | UD-Q3_K_XL/MiMo-V2-Flash-UD-Q3_K_XL-00001-of-00003.gguf |
| Q3_K_M | GGUF | 147.3 GB | — | Q3_K_M/MiMo-V2-Flash-Q3_K_M-00001-of-00003.gguf |
| IQ4_XS | GGUF | 164.4 GB | — | IQ4_XS/MiMo-V2-Flash-IQ4_XS-00001-of-00004.gguf |
| IQ4_NL | GGUF | 174.0 GB | — | IQ4_NL/MiMo-V2-Flash-IQ4_NL-00001-of-00004.gguf |
| Q4_0 | GGUF | 174.7 GB | — | Q4_0/MiMo-V2-Flash-Q4_0-00001-of-00004.gguf |
| Q4_K_S | GGUF | 175.4 GB | — | Q4_K_S/MiMo-V2-Flash-Q4_K_S-00001-of-00004.gguf |
| Q4_K_XL | GGUF | 177.7 GB | — | UD-Q4_K_XL/MiMo-V2-Flash-UD-Q4_K_XL-00001-of-00004.gguf |
| Q4_K_M | GGUF | 186.8 GB | — | Q4_K_M/MiMo-V2-Flash-Q4_K_M-00001-of-00004.gguf |
| Q4_1 | GGUF | 193.3 GB | — | Q4_1/MiMo-V2-Flash-Q4_1-00001-of-00004.gguf |
| Q5_K_S | GGUF | 212.5 GB | — | Q5_K_S/MiMo-V2-Flash-Q5_K_S-00001-of-00005.gguf |
| Q5_K_M | GGUF | 219.1 GB | — | Q5_K_M/MiMo-V2-Flash-Q5_K_M-00001-of-00005.gguf |
| Q5_K_XL | GGUF | 219.8 GB | — | UD-Q5_K_XL/MiMo-V2-Flash-UD-Q5_K_XL-00001-of-00005.gguf |
| Q6_K | GGUF | 253.5 GB | — | Q6_K/MiMo-V2-Flash-Q6_K-00001-of-00006.gguf |
| Q6_K_XL | GGUF | 263.3 GB | — | UD-Q6_K_XL/MiMo-V2-Flash-UD-Q6_K_XL-00001-of-00006.gguf |
| Q8_0 | GGUF | 328.2 GB | — | Q8_0/MiMo-V2-Flash-Q8_0-00001-of-00007.gguf |
| Q8_K_XL | GGUF | 354.2 GB | — | UD-Q8_K_XL/MiMo-V2-Flash-UD-Q8_K_XL-00001-of-00008.gguf |
| BF16 | GGUF | 617.7 GB | — | BF16/MiMo-V2-Flash-BF16-00001-of-00013.gguf |
Sizes are file sizes, not peak runtime memory.
Raw benchmarks
The inputs.
| Source | Task | Raw | Normalized |
|---|---|---|---|
| ZeroEval | AIME2025 | 94.1 | 94.1 |
| ZeroEval | GPQA | 83.7 | 83.7 |
| ZeroEval | HLE | 22.1 | 22.1 |
| ZeroEval | SWEBenchVerified | 73.4 | 73.4 |
From the Hugging Face model card
About this model.
From the Hugging Face model card