utOpenMOSS · Downloadable
MOSS-VL-Base-0408
Generates text for conversations and writing.
- Source listed
- Apr 7, 2026
- Parameters
- 11.3B
- 4-bit weights only
- ≈ 5.7 GB
- License
- apache-2.0
YOUR MEMORY BUDGET
Local AI models for 12GB VRAM.
Explore models by theoretical weight storage. This shortlist covers available downloadable language models with a known or name-derived parameter count.
10 GBleft for weights at 4-bit
12 GB budget − 2 GB reserve
12 GB budget − 2 GB reserve
The reserve is your assumption, not a measured overhead allowance. The default 2 GB may be too little for your runtime or context length. Results are not confirmed fits and have no verified speed rating. Weight estimates use decimal GB and do not include quantization overhead. For timed results on one 12GB machine, see measured speeds.
978 models within the weight budget
Largest parameter count first · not a quality rankingOpenMOSS · Downloadable
MOSS-VL-Base-0708
Generates text for conversations and writing.
- Source listed
- Jul 13, 2026
- Parameters
- 11.3B
- 4-bit weights only
- ≈ 5.7 GB
- License
- apache-2.0
OpenMOSS · Downloadable
MOSS-VL-Instruct-0408
Generates text for conversations and writing.
- Source listed
- Apr 7, 2026
- Parameters
- 11.3B
- 4-bit weights only
- ≈ 5.7 GB
- License
- apache-2.0
OpenMOSS · Downloadable
MOSS-VL-Instruct-0708
Generates text for conversations and writing.
- Source listed
- Jul 13, 2026
- Parameters
- 11.3B
- 4-bit weights only
- ≈ 5.7 GB
- License
- apache-2.0
OpenMOSS · Downloadable
MOSS-VL-Instruct-0708-NF4
Generates text for conversations and writing.
- Source listed
- Aug 11, 2026
- Parameters
- 11.3B
- 4-bit weights only
- ≈ 5.7 GB
- License
- apache-2.0
OpenMOSS · Downloadable
MOSS-VL-Realtime
Generates text for conversations and writing.
- Source listed
- Jul 14, 2026
- Parameters
- 11.3B
- 4-bit weights only
- ≈ 5.7 GB
- License
- apache-2.0
OpenMOSS · Downloadable
MOSS-VL-Realtime-NF4
Generates text for conversations and writing.
- Source listed
- Aug 7, 2026
- Parameters
- 11.3B
- 4-bit weights only
- ≈ 5.7 GB
- License
- apache-2.0
OpenMOSS · Downloadable
MOSS-VL-Realtime-SGLANG
Generates text for conversations and writing.
- Source listed
- Sep 6, 2026
- Parameters
- 11.3B
- 4-bit weights only
- ≈ 5.7 GB
- License
- apache-2.0
Meta · Downloadable
Llama-3.2-11B-Vision
Understands text and images and responds in text.
- Source listed
- Sep 18, 2024
- Parameters
- 11B
- 4-bit weights only
- ≈ 5.5 GB
- License
- llama3.2
Meta · Downloadable
Llama-3.2-11B-Vision-Instruct
Understands text and images and responds in text.
- Source listed
- Sep 18, 2024
- Parameters
- 11B
- 4-bit weights only
- ≈ 5.5 GB
- License
- llama3.2
Meta · Downloadable
Llama-Guard-3-11B-Vision
Understands text and images and responds in text.
- Source listed
- Sep 20, 2024
- Parameters
- 11B
- 4-bit weights only
- ≈ 5.5 GB
- License
- llama3.2
BAAI · Downloadable
OpenSeek-Mid-v1
Generates text for conversations and writing.
- Source listed
- Apr 30, 2026
- Parameters
- 10.6B
- 4-bit weights only
- ≈ 5.3 GB
- License
- Not disclosed
Z.ai · Downloadable
GLM-4.6V-Flash
Understands text and images and responds in text.
- Source listed
- Dec 7, 2025
- Parameters
- 10.3B
- 4-bit weights only
- ≈ 5.1 GB
- License
- mit
Z.ai · Downloadable
Glyph
Understands text and images and responds in text.
- Source listed
- Oct 25, 2025
- Parameters
- 10.3B
- 4-bit weights only
- ≈ 5.1 GB
- License
- mit
Z.ai · Downloadable
UI2Code_N
Designed for coding and software tasks. Understands text and images and responds in text.
- Source listed
- Nov 11, 2025
- Parameters
- 10.3B
- 4-bit weights only
- ≈ 5.1 GB
- License
- mit
Z.ai · Downloadable
WebVIA-Agent
Understands text and images and responds in text.
- Source listed
- Nov 12, 2025
- Parameters
- 10.3B
- 4-bit weights only
- ≈ 5.1 GB
- License
- mit
OpenBMB · Downloadable
cpm-ant-10b
Generates text for conversations and writing.
- Source listed
- Jan 15, 2023
- Parameters
- 10B
- 4-bit weights only
- ≈ 5 GB
- License
- Not disclosed
OpenBMB · Downloadable
cpm-bee-10b
Generates text for conversations and writing.
- Source listed
- May 26, 2023
- Parameters
- 10B
- 4-bit weights only
- ≈ 5 GB
- License
- Not disclosed
StepFun · Downloadable
Step3-VL-10B
Understands text and images and responds in text.
- Source listed
- Jan 13, 2026
- Parameters
- 10B
- 4-bit weights only
- ≈ 5 GB
- License
- apache-2.0
StepFun · Downloadable
Step3-VL-10B-Base
Understands text and images and responds in text.
- Source listed
- Jan 13, 2026
- Parameters
- 10B
- 4-bit weights only
- ≈ 5 GB
- License
- apache-2.0
Datalab · Downloadable
lift
Understands text and images and responds in text.
- Source listed
- Jun 19, 2026
- Parameters
- 9.7B
- 4-bit weights only
- ≈ 4.8 GB
- License
- openrail
OpenBMB · Downloadable
MiniCPM-SALA
Generates text for conversations and writing.
- Source listed
- Feb 11, 2026
- Parameters
- 9.5B
- 4-bit weights only
- ≈ 4.7 GB
- License
- apache-2.0
Black Forest Labs · Downloadable
flux-3-image-klein-pup
Understands text and images and responds in text.
- Source listed
- Oct 14, 2026
- Parameters
- 9.4B
- 4-bit weights only
- ≈ 4.7 GB
- License
- apache-2.0
Cloudflare · Downloadable
clef-flash
Understands text and images and responds in text.
- Source listed
- Sep 30, 2026
- Parameters
- 9.4B
- 4-bit weights only
- ≈ 4.7 GB
- License
- apache-2.0
What changes with a 12GB budget?
With 2 GB set aside, 10 GB remains for weights. At 4-bit that allows up to 20 billion parameters in the theoretical calculation. An 8-billion-parameter model alone needs approximately 4 GB at this precision, before its conversation cache and runtime.
Why might a listed model still fail to load?
Its actual weight format may need more space, the runtime may not support its architecture, or the conversation cache may exceed your reserve. Vision models can also need space for image processing. Check the original model card and measure your exact setup before relying on it.
Read the memory guide and technical sources · Compare two models