deepseek-v3

deepseek-v3.1

DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.

tools thinking cloud 671b

146.7K Pulls 8 Tags Updated 1 month ago

A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

671b

2.7M Pulls 5 Tags Updated 10 months ago

huihui_ai/deepseek-v3

A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

7,115 Pulls 2 Tags Updated 9 months ago

huihui_ai/deepseek-v3-abliterated

A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

671b

2,728 Pulls 5 Tags Updated 7 months ago

milkey/deepseek-v3-UD

(Unsloth Dynamic Quants) A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

2,012 Pulls 3 Tags Updated 9 months ago

huihui_ai/deepseek-v3-pruned

DeepSeek-V3-Pruned-Coder-411B is a pruned version of the DeepSeek-V3 reduced from 256 experts to 160 experts, The pruned model is mainly used for code generation.

411b

1,261 Pulls 5 Tags Updated 7 months ago

8b-wraith/deepseek-v3-0324

deepseek-v3-0324-Quants. - Q2_K is the lowest here - quantized = round((original - zero_point) / scale)

1,009 Pulls 1 Tag Updated 7 months ago

MFDoom/deepseek-v3-tool-calling

tools 671b

900 Pulls 2 Tags Updated 9 months ago

lwk/v3

ollama run deepseek-v3

tools

800 Pulls 1 Tag Updated 9 months ago

xiaowangge/deepseek-v3-qwen2.5

This model has been developed based on DistilQwen2.5-DS3-0324-Series.

tools 32b

789 Pulls 7 Tags Updated 6 months ago

sunny-g/deepseek-v3-0324

dynamic quants from unsloth, merged

291 Pulls 1 Tag Updated 7 months ago

org/deepseek-v3-fast

Single file version with (Dynamic Quants) A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

90 Pulls 4 Tags Updated 8 months ago

huihui_ai/deepseek-v3.1

This is not the ablation version. DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode.

tools thinking 671b

75 Pulls 3 Tags Updated 2 months ago

pdevine/deepseek-v3.1

cloud

26 Pulls 2 Tags Updated 1 month ago

lucataco/deepseek-v3-64k

A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

19 Pulls 1 Tag Updated 10 months ago

clore/deepseek-v3.1

12 Pulls 1 Tag Updated 2 months ago

ukjin/Qwen3-30B-A3B-Thinking-2507-Deepseek-v3.1-Distill

This model is a distilled version of Qwen/Qwen3-30B-A3B-Instruct designed to inherit the reasoning and behavioral characteristics of its much larger teacher model, deepseek-ai/DeepSeek-V3.1.

tools thinking 4b

642 Pulls 2 Tags Updated 2 months ago

haghiri/DeepSeek-V3-0324

Merged Unsloth's Dynamic Quantization

1,339 Pulls 1 Tag Updated 7 months ago

lsm03624/deepseek-r1

DeepSeek-R1-0528 仍然使用 2024 年 12 月所发布的 DeepSeek V3 Base 模型作为基座，但在后训练过程中投入了更多算力，显著提升了模型的思维深度与推理能力。这个8B精馏版本编程能力都爆表！

thinking

797 Pulls 1 Tag Updated 5 months ago

chsword/DeepSeek-V3

tools

616 Pulls 1 Tag Updated 10 months ago