utMistral AI · Downloadable
Mistral-7B-Instruct-v0.2
Generates text for conversations and writing.
- Source listed
- Dec 11, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
YOUR MEMORY BUDGET
Local AI models for 24GB VRAM.
Explore models by theoretical weight storage. This shortlist covers available downloadable language models with a known or name-derived parameter count.
22 GBleft for weights at 4-bit
24 GB budget − 2 GB reserve
24 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.
1,130 models within the weight budget
Largest parameter count first · not a quality rankingMistral AI · Downloadable
Mistral-7B-Instruct-v0.3
Generates text, according to its published model architecture (MistralForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.
- Source listed
- May 22, 2024
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
Mistral AI · Downloadable
Mistral-7B-v0.1
Generates text for conversations and writing.
- Source listed
- Sep 20, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
Mistral AI · Downloadable
Mistral-7B-v0.3
Generates text, according to its published model architecture (MistralForCausalLM). No task is listed on Hugging Face, so check the model card for intended use.
- Source listed
- May 22, 2024
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenBMB · Downloadable
Eurus-7b-kto
Generates text for conversations and writing.
- Source listed
- Apr 1, 2024
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenBMB · Downloadable
Eurus-7b-sft
Generates text for conversations and writing.
- Source listed
- Apr 1, 2024
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenBMB · Downloadable
RLAIF-V-7B
Generates text for conversations and writing.
- Source listed
- May 19, 2024
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenBMB · Downloadable
RLPR-Qwen2.5-7B-Base
Generates text for conversations and writing.
- Source listed
- Jun 22, 2025
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenMOSS · Downloadable
character-llm-beethoven-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
character-llm-caesar-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
character-llm-cleopatra-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- cc-by-nc-4.0
OpenMOSS · Downloadable
character-llm-hermione-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
character-llm-martin-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
character-llm-newton-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
character-llm-socrates-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
character-llm-spartacus-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
character-llm-voldemort-7b-wdiff
Generates text for conversations and writing.
- Source listed
- Nov 14, 2023
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
Game-RL-Qwen2.5-VL-7B
Understands text and images and responds in text.
- Source listed
- May 27, 2025
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenMOSS · Downloadable
Llama-2-7B-MHA-d_kv_256
Generates text for conversations and writing.
- Source listed
- Mar 3, 2025
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenMOSS · Downloadable
Llama-2-7B-MLA-d_kv_16
Generates text for conversations and writing.
- Source listed
- Mar 3, 2025
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenMOSS · Downloadable
Llama-2-7B-MLA-d_kv_32
Generates text for conversations and writing.
- Source listed
- Mar 3, 2025
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenMOSS · Downloadable
Llama-2-7B-MLA-d_kv_64
Generates text for conversations and writing.
- Source listed
- Mar 3, 2025
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- apache-2.0
OpenMOSS · Downloadable
llama2-7B-d_kv_16-refactor
Generates text for conversations and writing.
- Source listed
- Jun 17, 2025
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
OpenMOSS · Downloadable
llama2-7B-d_kv_32-refactor
Generates text for conversations and writing.
- Source listed
- Jun 17, 2025
- Parameters
- 7B
- 4-bit weights only
- ≈ 3.5 GB
- License
- Not disclosed
What changes with a 24GB budget?
With 2 GB set aside, 22 GB remains for weights. At 4-bit that allows up to 44 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