vllm-project/Kimi-K3.1-Pro-pruned75
archived · low-signalmodel
Image text to text model from vllm-project (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 738B total (224 experts, 16 per token). Context: 1M tokens (config). License: kimi-k3.1-pro.
- Org
- vllm-project
- Released
- 2026-11-05
- Status
- archived · low-signal
- License
- kimi-k3.1-pro (restricted)
- Parameters
- 737.6B
- Architecture
- MoE · 224 experts, 16/token · KimiK31ForConditionalGeneration
- Context
- 1,048,576 tokens (config.json)
- Task
- image-text-to-text · vision
- Signal
- 20 HF likes · 254 downloads
- Archived
- 2026-11-09 (no longer listed by the collector; stats as last seen)
Runs on
estimate from params; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~444.5 GB (estimated size): no listed device
- Q8 ~785.6 GB (estimated size): no listed device
- FP16 ~1477.1 GB (estimated size): no listed device
Links
First published here 2026-11-05 · updated 2026-11-09 · markdown · JSON