Downloads · 30 days
32
4% of all-time downloads
bullerwins/DeepSeek-V4-Flash-0731-exl3-3.82bpw
DeepSeek-V4-Flash-0731-exl3-3.82bpw is a text generation model from bullerwins. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as mit.
This checkpoint is part of Bullerwins' final DeepSeek-V4-Flash-0731 EXL3 ladder. The chart compares our retained quants (blue) with Turboderp's ladder (orange): mean token KLD is shown against decoder bitrate and tota…
Downloads · 30 days
32
4% of all-time downloads
All-time downloads
782
Public
Parameters
75.3B
151 GB on disk
Likes
2
Public
Click a slice to open those files.
.safetensors151 GB · 100%
How the weights are stored.
I1674.4B · 99%
From the Hugging Face model README
This checkpoint is part of Bullerwins' final DeepSeek-V4-Flash-0731 EXL3 ladder. The chart compares our retained quants (blue) with Turboderp's ladder (orange): mean token KLD is shown against decoder bitrate and total weight size, followed by perplexity against size. Lower is better; the dotted lines mark reference self-noise or official-checkpoint perplexity. All measurements used formatted OpenWebText, 10 × 2048 tokens, and the official FP8/MXFP4 checkpoint as reference.

DeepSeek-V4-Flash-0731 is the official release of DeepSeek-V4-Flash, superseding the preview version, with substantially enhanced agentic capabilities. It has the same model structure as DeepSeek-V4-Flash-DSpark, i.e. it comes with a speculative decoding module attached.
DeepSeek-V4-Flash-0731 outperforms DeepSeek-V4-Pro (Preview) on benchmarks listed below despite its far smaller activated parameter count, and is broadly competitive with the strongest proprietary models available.
<div align="center">| Benchmark | DeepSeek-V4-Flash-0731 | DeepSeek-V4-Flash (Preview) | DeepSeek-V4-Pro (Preview) | GLM-5.2 | Opus-4.8 |
|---|---|---|---|---|---|
| Terminal Bench 2.1 | 82.7 | 61.8 | 72.1 | 81.0 | 85.0 |
| NL2Repo | 54.2 | 39.4 | 38.5 | 48.9 | 69.7 |
| Cybergym | 76.7 | 38.7 | 52.7 | - | 83.1 |
| DeepSWE | 54.4 | 7.3 | 12.8 | 46.2 | 58.0 |
| Toolathlon-Verified | 70.3 | 49.7 | 55.9 | 59.9 | 76.2 |
| Agents' Last Exam | 25.2 | 15.8 | 16.5 | 23.8 | 25.7 |
| AutomationBench Public | 25.1 | 10.8 | 12.8 | 12.9 | 27.2 |
| DSBench-FullStack † | 68.7 | 37.0 | 41.8 | 61.8 | 71.6 |
| DSBench-Hard † | 59.6 | 25.8 | 31.1 | 54.5 | 71.7 |
Notes:
max reasoning effort level with temperature = 1.0, top_p = 0.95.This release does not include a Jinja-format chat template. Instead, we provide a dedicated encoding folder with Python scripts and test cases demonstrating how to encode messages in OpenAI-compatible format into input strings for the model, and how to parse the model's text output. Please refer to the encoding folder for full documentation.
The reasoning_effort parameter now supports three levels — low, high, and max — which control how much deliberation the model spends before answering.
A brief example:
from encoding_dsv4 import encode_messages, parse_message_from_completion_text
messages = [
{"role": "user", "content": "hello"},
{"role": "assistant", "content": "Hello! I am DeepSeek.", "reasoning_content": "thinking..."},
{"role": "user", "content": "1+1=?"}
]
# messages -> string
prompt = encode_messages(messages, thinking_mode="thinking", reasoning_effort="max")
# string -> tokens
import transformers
tokenizer = transformers.AutoTokenizer.from_pretrained("deepseek-ai/DeepSeek-V4-Flash-0731")
tokens = tokenizer.encode(prompt)
DSpark speculative decoding is enabled with a single flag — add --speculative-config with method: dspark to your vLLM launch command:
--speculative-config '{"method":"dspark","num_speculative_tokens":7,"draft_sample_method":"greedy"}'
For example, the command below serves the model with vLLM on a single 4×GB300 node. See the vLLM recipe for detailed instructions and other hardware configurations.
vllm serve deepseek-ai/DeepSeek-V4-Flash-0731 \
--trust-remote-code --kv-cache-dtype fp8 --block-size 256 \
--data-parallel-size 4 --enable-expert-parallel \
--moe-backend deep_gemm_mega_moe \
--attention-config '{"use_fp4_indexer_cache": true}' \
--speculative-config '{"method":"dspark","num_speculative_tokens":7,"draft_sample_method":"greedy"}'
Enable DSpark with --speculative-algorithm DSPARK and do not set a separate --speculative-draft-model-path as the target and draft weights therefore come from the same checkpoint.
See the SGLang cookbook for detailed instructions, benchmarks and other hardwares configurations.
sglang serve \
--trust-remote-code \
--model-path deepseek-ai/DeepSeek-V4-Flash-0731 \
--tp 4 \
--moe-runner-backend flashinfer_mxfp4 \
--speculative-algorithm DSPARK \
--mem-fraction-static 0.90 \
--chunked-prefill-size 4096 \
--swa-full-tokens-ratio 0.1 \
Please refer to the inference folder for detailed instructions on running DeepSeek-V4 locally, including model weight conversion and interactive chat demos.
For local deployment, we recommend setting the sampling parameters to temperature = 1.0, with top_p = 0.95 for agentic scenarios and top_p = 1.0 otherwise. For the high and max reasoning effort levels, we recommend a maximum output length of 384K tokens.
This repository and the model weights are licensed under the MIT License.
@misc{deepseekai2026deepseekv4,
title={DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence},
author={DeepSeek-AI},
year={2026},
}
If you have any questions, please raise an issue or contact us at [email protected].