No longer available on HF due to storage restrictions: archived here
See MiniMax-M2.7 in action: demonstration videos
Tested with an M3 Ultra 512 GiB using Inferencer app
- Text inference: ~37.45 tokens/s @ 1000 tokens ~161 GiB (debug build)
Q6-INF uses the data-agnostic INF method tuned to yield maximum general accuracy within a 192 GiB memory budget
| Quantization (bpw) | Perplexity | Token Accuracy | Missed Divergence |
|---|---|---|---|
| Q4.5 | 1.27343 | 92.40% | 24.73% |
| Q6-INF | 1.20312 | 97.40% | 13.92% |
| Q6.5 | 1.21093 | 96.85% | 11.74% |
| Q9 | 1.20312 | 97.50% | 9.95% |
| Base | 1.20312 | 100.0% | 0.000% |
- Perplexity: Measures the confidence for predicting base tokens (lower is better)
- Token Accuracy: The percentage of correctly generated base tokens
- Missed Divergence: Measures severity of misses; how much the token was missed by
Quantized with a modified version of MLX
For more details see our demonstration videos or visit MiniMax-M2.7.
Model tree for inferencerlabs/MiniMax-M2.7-MLX-Q6-INF
Base model
MiniMaxAI/MiniMax-M2.7