Text / deepseek-ai

DeepSeek V4 Pro

DeepSeek V4 Pro is a preview Mixture-of-Experts text-generation model from the DeepSeek-V4 series, released under the MIT license with Safetensors and Transformers compatibility. It has 1.6T total parameters with 49B activated and supports a context length of one million tokens. The model card describes a hybrid attention architecture combining Compressed Sparse Attention and Heavily Compressed Attention, Manifold-Constrained Hyper-Connections, and Muon optimizer training. It was pre-trained on more than 32T tokens and post-trained with a two-stage paradigm involving domain-specific SFT and RL with GRPO followed by on-policy distillation. A DeepSeek-V4-Pro-Max mode is described as offering maximum reasoning effort.

EstimatedRepository opened Apr 22, 2026Source checked 9/24/2026Version: b5968e91

LOCALRENTED GPUOPEN WEIGHTS

Freshness 51 out of 100. Repository age: 176 days old, Apr 22, 2026. Measures age, not quality.

Popularity 79 out of 100. 383,720 downloads in 30 days; 5,673 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.

Arena Open Source Text: rank 11 of 225. Ranked model: deepseek-v4-pro. Export dated Oct 13, 2026. Source checked Oct 15, 2026. Human preference ranking, not hardware suitability.

Category rankings & sources

Each position belongs to a separate Arena category. These are human preference rankings—not hardware recommendations. We do not combine the categories into one score.

Text#11of 225 open models

deepseek-v4-prodeepseekLicense: MIT

Export dated Oct 13, 2026Checked: Oct 15, 2026Overall board: #64

View official Text board

Web coding#20of 55 open models

deepseek-v4-prodeepseekLicense: MIT

Export dated Oct 13, 2026Checked: Oct 15, 2026Overall board: #71

View official Web coding board

Rankings reflect the dated export; live Arena boards can change. Open-source positions use Arena’s open-model filter. The overall position also includes the other models on that category board. Availability of weights and license terms still need to be checked separately.

What do these scores mean?

Two ways to compare models. Neither score measures answer quality, benchmark performance, or whether a model fits your machine.

Freshness · release age

A new release starts at 100. The score halves every 180 days: 100 × 2^(−age in days / 180). When the release date is unavailable, a repository date is a fallback—not proof of when the model was released. Missing dates stay unscored.

Check the date source

Popularity · adoption

Downloads over the past 30 days carry 80% of the score; lifetime likes carry 20%. Both use a logarithmic scale, capped at 10 million downloads and 100,000 likes. If likes are unavailable, downloads carry all the weight. Missing download statistics stay unscored.

round(80 × min(1, log10(1 + downloads) / 7) + 20 × min(1, log10(1 + likes) / 5))

Check the statistics source

Scores as of Oct 15, 2026. Statistics checked Oct 15, 2026. Statistics are marked for refresh after 72 hours.Full scoring method

At a glance

Parameters: 1598.84B
Architecture: deepseek_v4
Context length: 1,048,576
License: mit
Disk space: 805.3 GB
Software: Publisher PyTorch reference

View the model source

MY HARDWARE

Will it run on your machine?

Save your machine to see a personalized rating and its reasoning.

Add my hardware

Best for

Long-context text generation
Coding tasks
Reasoning tasks
Agentic tasks
Knowledge-intensive applications

Tradeoffs

Only a preview version is presented
Detailed compatibility and runtime requirements are not specified in the supplied facts
Performance figures are self-reported in the model card and not independently verified
The model card is truncated, so complete details are unavailable
No information is provided about ethical, safety, or bias considerations

Ways to run it

Publisher PyTorch reference · Linux / rented GPU

Repository-specific command in publisher documentation; confirm dependencies and hardware in the source.

torchrun --nproc-per-node ${MP} generate.py --ckpt-path ${SAVE_PATH} --config ${CONFIG} --interactive

Official instructions

Explore its uses

SAFETENSORSTEXT-GENERATION

Keep exploring

Text

Prism ML

Bonsai 2 · 27B

A compressed 27B-class reasoning model with publisher-provided GGUF packs and a dedicated llama.cpp fork for CUDA, Metal, and CPU.

Freshness 90 out of 100. Release age: 28 days old, Sep 17, 2026. Measures age, not quality.

MODEL SIZE27.36B

LOCALRENTED GPU

Released Sep 17, 2026

Text

DeepSeek

DeepSeek V4.1 Flash

A multimodal reasoning model with a compressed key-value cache, published as open weights.

Freshness 87 out of 100. Repository age: 35 days old, Sep 10, 2026. Measures age, not quality.

Arena Open Source Text: rank 5 of 225. Ranked model: deepseek-v4.1-flash-max. Export dated Oct 13, 2026. Source checked Oct 15, 2026. Human preference ranking, not hardware suitability.+3

LOCALRENTED GPU

Repository opened Sep 10, 2026

Text

Qwen

Qwen3.8

Explore 4 versions of Qwen3.8. Compare model sizes, packages, hardware requirements, and available setups.

4 versionsCompare sizes & formats

LOCALRENTED GPU

Compare versions →