Open models: what you need to run them

Newest models of each family first. Models marked older have been superseded but are still widely downloaded. Last checked against Hugging Face 20 Oct 2026, 2:26 pm AEST.

ModelFamilyReleasedParamsQ4 weights
Can I run DeepSeek V4.1 Flash 1017 (MoE)? new
DeepSeek
17 Oct 2026
763.21B (MoE, active not published)
462.7 GB
Can I run Muse Spark Open 816B-A42B (MoE)? new
Meta
15 Oct 2026
822.04B (42B active, MoE)
498.4 GB
Can I run Hy Memory 30B A3B (MoE)? new
Tencent
15 Oct 2026
30.06B (3B active, MoE)
18.2 GB
Can I run Ling 3.1 flash (MoE)? new
Ling
13 Oct 2026
572.12B (25B active, MoE)
346.8 GB
Can I run LFM2.5 24B-A2B (MoE)? new
Liquid
12 Oct 2026
23.97B (2B active, MoE)
14.5 GB
Can I run Nemotron 3.5 Bolt 74B-A9B (MoE)? new
NVIDIA
7 Oct 2026
74.04B (9B active, MoE)
44.9 GB
Can I run MiniCPM V 4.7 35B A3B (MoE)? new
MiniCPM
7 Oct 2026
35.21B (3B active, MoE)
21.3 GB
Can I run Beam 1.0 501B A23B (MoE)? new
Reflection
7 Oct 2026
501.27B (23B active, MoE)
303.9 GB
Can I run MiniCPM V 4.7 1B? new
MiniCPM
6 Oct 2026
1.55B
0.9 GB
Can I run DeepSeek V4.1 Flash (MoE)? new
DeepSeek
10 Sept 2026
763.21B (MoE, active not published)
462.7 GB
Can I run GLM-5.3-Flash (MoE)? new
GLM
25 Aug 2026
321.32B (18B active, MoE)
194.8 GB
Can I run GLM 5.3 (MoE)? new
GLM
25 Aug 2026
753.33B (MoE, active not published)
456.7 GB
Can I run Qwen3.8 Flash-Next (MoE)? new
Qwen
24 Aug 2026
180B (6B active, MoE)
109.1 GB
Can I run Qwen3.8 27B?
Qwen
5 Aug 2026
27.78B
16.8 GB
Can I run Nemotron 3.5 Lightning 30B-A3B (MoE)?
NVIDIA
1 Aug 2026
31.58B (3B active, MoE)
19.1 GB
Can I run Gemma 4 12B?
Gemma
23 May 2026
11.96B
7.3 GB
Can I run Qwen3.6 35B-A3B (MoE)?
Qwen
15 Apr 2026
35.95B (3B active, MoE)
21.8 GB
Can I run Mistral Medium 3.5 128B?
Mistral
31 Mar 2026
127.7B
77.4 GB
Can I run Gemma 4 31B?
Gemma
11 Mar 2026
31.27B
19.0 GB
Can I run Gemma 4 26B-A4B (MoE)?
Gemma
11 Mar 2026
25.81B (3.8B active, MoE)
15.6 GB
Can I run Gemma 4 E4B?
Gemma
2 Mar 2026
8B
4.8 GB
Can I run Qwen3.5 9B?
Qwen
27 Feb 2026
9.65B
5.9 GB
Can I run Qwen3.5 4B?
Qwen
27 Feb 2026
4.66B
2.8 GB
Can I run Qwen3.5 122B-A10B (MoE)?
Qwen
24 Feb 2026
125.09B (10B active, MoE)
75.8 GB
Can I run GLM-4.7-Flash 30B-A3B (MoE)?
GLM
19 Jan 2026
31.22B (3B active, MoE)
18.9 GB
Can I run gpt-oss 20B (MoE)? older
OpenAI
4 Aug 2025
21B (3.6B active, MoE)
11.2 GB
Can I run gpt-oss 120B (MoE)? older
OpenAI
4 Aug 2025
117B (5.1B active, MoE)
62.2 GB
Can I run Qwen 3 4B? older
Qwen
27 Apr 2025
4.02B
2.4 GB
Can I run Qwen 3 8B? older
Qwen
27 Apr 2025
8.19B
5.0 GB
Can I run Qwen 3 32B? older
Qwen
27 Apr 2025
32.76B
19.9 GB
Can I run Llama 3.1 8B Instruct? older
Llama
18 July 2024
8.03B
4.9 GB
Can I run Mistral 7B Instruct v0.3? older
Mistral
22 May 2024
7.25B
4.4 GB