- Status
- Verified
- Base model
- …
- Lineage
- …
- Unchanged from base
- …
- Trending
- #1
- Downloads, 30 days
- 170k
- Weights
- 510 GB
- Revision
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.
At a glance
- Task
- Vision language
- Input
- image, text
- Output
- text
- Parameters
- 763B, 384 experts, 6 active
- Architecture
- DeepSeek V41
- Context
- 1M tokens
- Precision
- INT8 73%, FP8 27%
- Format
- Safetensors
- Library
- transformers
- License
- MITCommercial use
- Released
- Oct 2026
- Updated
- Oct 2026
- Likes
- 2,846
- Downloads, all time
- 73,215
Architecture
- Layers
- 40
- Hidden size
- 5,120
- Attention
- 64 heads, multi query, 1 KV head
- Experts
- 384 total, 6 active per token
- Vocabulary
- 129,280
- Positions
- 1,048,576
- RoPE theta
- 10,000
- Tied embeddings
- No
- Quantization
- fp8
- Vision encoder
- deepseek_v41_vision
Benchmarks
5 results, self reported by the authors.
| Benchmark | Score | Date | Evidence |
|---|---|---|---|
| cais/hlehle | 42.6 | Oct 2026 | Reported by source |
| cais/hlehle | 44 | Oct 2026 | Reported by source |
| harborframework/terminal-benchterminalbench_4 | 42.3 | Oct 2026 | Reported by source |
| datacurve/deep-swedeep_swe | 77 | Oct 2026 | Reported by source |
| agents-last-exam/agents-last-examoverall-pass-rate | 35.4 | Oct 2026 | Reported by source |
Family
Models built on DeepSeek-V4.1-Flash-1017.
Run it
Loads with Transformers AutoModelForCausalLM, pinned to the indexed revision.
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-V4.1-Flash-1017", revision="1bae0e1088993b9b82de08f59de5be5a34e42170")Through the hub: the same tools, each file from a source that is up (Hugging Face, ModelScope, IPFS), at this revision. The second line checks every file against its address.
export HF_ENDPOINT=https://gethologram.ai
cd "$(hf download deepseek-ai/DeepSeek-V4.1-Flash-1017 --quiet)" && curl -s $HF_ENDPOINT/deepseek-ai/DeepSeek-V4.1-Flash-1017/resolve/main/SHA256SUMS | sha256sum -c --quietHosted inference, live now
Spaces
Used in 14 Spaces.
Read the full model card
DeepSeek-V4.1-Flash-1017
DeepSeek-V4.1
Introduction
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. Alongside these gains, the model reduces task execution time and output length, enabling users to complete tasks faster and at lower cost and advancing the performance–cost Pareto frontier across multiple agent harnesses.
| Benchmark | DS-V4.1-Flash-1017 | DS-V4.1-Flash | K3 | GLM-5.3 | GPT-5.6 Sol | Opus 5 |
|---|---|---|---|---|---|---|
| HLE w/o tools | 42.6 (44.0†) | 36.8 (39.1†) | 43.5 | 42.0† | 44.5 | 56.3 |
| Terminal-Bench 4.0 | 42.3 | 31.2 | 12.6 | 37.9 | 39.9 | 51.8 |
| DeepSWE v1.1 | 77.0 | 74.2 | 67.5 | 66.9 | 73.0 | 74.0 |
| ProgramBench | 29.1 | 20.3 | 17.5 | 19.0 | 23.0 | 37.0 |
| NL2Repo-Bench | 68.2 | 65.4 | 58.0 | 58.0 | 56.8 | 75.3 |
| Automation-Bench v1.0.6 | 60.5 | 54.8 | 46.7 | 48.8 | 45.8 | 50.3 |
| Agents’ Last Exam | 35.4 | 31.8 | 27.6 | 28.5 | 26.7 | 28.6 |
| Apex Bench v1.1 | 72.1 | 64.6 | 47.8 | 56.6 | 54.3 | 65.8 |
| Chartography (w/tools) | 83.0 | 78.9 | 68.1 | — | 79.9 | 84.0 |
| BabyVision (w/tools) | 92.5 | 89.6 | 85.7 | — | 88.9 | 94.1 |
| ZeroBench-main (w/tools) | 51.0 | 49.0 | 41.0 | — | 53.0 | 52.0 |
† denotes the text-only subset of HLE. We use the Claude Code harness for Terminal-Bench 4.0 and the Mini-SWE-V2 harness for DeepSWE v1.1.
Prompt Encoding
This release does not include a Jinja-format chat template. The encoding folder contains a self-contained Python reference implementation (encoding.py) with test cases for multi-turn conversations, tool calling, thinking mode, numeric reasoning effort, mid-conversation system messages, and interleaved image content.
For production use, we additionally release deepseek-recipe, a set of Rust libraries with Python bindings that provides the same prompt format as a maintained, protocol-aware toolkit. It converts Messages, Chat Completions, and Responses API requests into the Conversation format, encodes them into DeepSeek V4 and V4.1 prompts or token IDs, and parses model output back into complete or streamed responses — covering thinking, tool calls, images, and generation settings. Model inference, tool execution, and HTTP transport are left to the caller.
Minimal Inference
Please refer to the inference folder for instructions on weight conversion and running inference locally.
Recommended sampling parameters:
| Parameter | Value |
|---|---|
temperature |
1.0 |
top_p |
1.0 |
context_window |
1M tokens |
max_tokens |
≥ 256K |
License
This repository and the model weights are licensed under the MIT License.
Citation
@misc{deepseekai2026deepseekv41flash,
title={DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression},
author={DeepSeek-AI},
year={2026},
}
Contact
If you have any questions, please raise an issue or contact us at service@deepseek.com.
Derived on Oct 25, 2026 from Hugging Face at revision 1bae0e10, README.md , config.json .
83 files, 510 GB. Every download is checked against its address.
