deepseek-ai/DeepSeek-V4.1-Flash-1017
main boardmodel
DeepSeek-V4.1-Flash-1017 is a minor update to DeepSeek-V4.1-Flash, delivering substantial improvements across reasoning, coding, agentic tasks, and visual benchmarks. Mixture-of-experts, 763B total (384 experts, 6 per token). Context: 1M tokens (config). License: mit.
- Org
- deepseek-ai
- Released
- 2026-10-17
- Status
- main board
- License
- mit (permissive)
- Parameters
- 763.2B
- Architecture
- MoE · 384 experts, 6/token · DeepseekV41ForCausalLM
- Context
- 1,048,576 tokens
- Task
- image-text-to-text · vision
- Signal
- 2906 HF likes · 85275 downloads · 832 X likes · 3 X post(s)
Runs on
mixed: real files + estimate; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~502.4 GB: no listed device
- Q8 ~510.1 GB: no listed device
- FP16 ~1528.1 GB (estimated size): no listed device
Links
- Hugging Face
- Ollama
- MLX: mlx-community/DeepSeek-V4.1-Flash-1017-oQ2.5e-mtp
- MLX: mlx-community/DeepSeek-V4.1-Flash-1017-oQ2.5e-ngram-mtp
- MLX: mlx-community/DeepSeek-V4.1-Flash-1017-oQ2e-mtp
- MLX: mlx-community/DeepSeek-V4.1-Flash-1017-oQ2e-ngram-mtp
- MLX: mlx-community/DeepSeek-V4.1-Flash-1017-oQ3e-mtp
- MLX: mlx-community/DeepSeek-V4.1-Flash-1017-oQ3e-ngram-mtp
- MLX: mlx-community/DeepSeek-V4.1-Flash-1017-oQ4e-mtp
- MLX: mlx-community/DeepSeek-V4.1-Flash-1017-oQ4e-ngram-mtp
- GGUF: unsloth/DeepSeek-V4.1-Flash-1017-GGUF
First published here 2026-10-17 · updated 2026-10-17 · markdown · JSON