internlm/AdvancedMathBench-AutoVerifier
main boardmodel
AutoVerifier is a learned grader, not a formal proof checker, and can make errors. Mixture-of-experts, 36B total (256 experts, 8 per token). Context: 256K tokens (config).
- Org
- internlm
- Released
- 2026-09-29
- Status
- main board
- License
- unknown
- Parameters
- 36.0B
- Architecture
- MoE · 256 experts, 8/token · Qwen3_5MoeForConditionalGeneration
- Context
- 262,144 tokens
- Task
- image-text-to-text · vision
- Signal
- 1 HF likes · 326 downloads
Runs on
mixed: real files + estimate; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~23.3 GB (estimated size): 32GB GPU (5090), Mac 36GB, Mac 48GB, CPU only, 32GB RAM (slow)
- Q8 ~39.9 GB (estimated size): 48GB GPU (RTX 6000 Ada / A6000), Mac 64GB (M1 Max), Mac 96GB, CPU only, 64GB RAM (slow)
- FP16 ~74.7 GB: 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)
Links
First published here 2026-09-29 · updated 2026-09-29 · markdown · JSON