Runs Perfectly (S/A/B)
76
With Offloading (C/D)
36
Cannot Run (F)
33
S
Excellent (67 models)
Runs great with room for large context windows
Granite 3.0 3B-A800M
IBM
3.4BChat / LLMQ4_K_M2.4GB
Llama 3.2 3B Instruct
Meta
3.2BChat / LLMQ5_K_M2.7GB
Qwen 2.5 3B
Alibaba
3BChat / LLMQ4_K_M2.5GB
Qwen 2.5 Coder 3B
Alibaba
3BCodingQ4_K_M2.5GB
Falcon 3 3B
TII
3BChat / LLMQ4_K_M2.4GB
StableLM Zephyr 3B
Stability AI
3BChat / LLMQ4_K_M2.1GB
Rocket 3B
Pansophic
3BChat / LLMQ4_K_M2.1GB
StarCoder2 3B
BigCode
3BCodingQ4_K_M2.3GB
Stable Code 3B
Stability AI
3BCodingQ4_K_M2.1GB
PaliGemma 3B
3BMultimodalQ4_K_M2.5GB
Gemma 2 2B
2.6BChat / LLMQ4_K_M2.1GB
EXAONE 3.5 2.4B
LG AI
2.4BChat / LLMQ4_K_M2.0GB
Qwen2-VL 2B
Alibaba
2.2BMultimodalQ8_02.0GB
CodeGemma 2B
2BCodingQ4_K_M2.0GB
MiniCPM-V 2.6
OpenBMB
2BMultimodalQ4_K_M2.1GB
Granite 3.3 2B
IBM
2BChat / LLMQ4_K_M1.9GB
Moondream 2
Moondream
1.8BMultimodalQ4_K_M1.5GB
SmolLM2 1.7B
HuggingFace
1.7BChat / LLMQ8_02.2GB
Qwen 2.5 1.5B
Alibaba
1.5BChat / LLMQ8_02.3GB
DeepSeek R1 Distill 1.5B
DeepSeek
1.5BChat / LLMQ8_02.3GB
Qwen 2.5 Coder 1.5B
Alibaba
1.5BCodingQ8_02.3GB
Yi Coder 1.5B
01.AI
1.5BCodingQ8_02.0GB
DeepSeek Coder 1.3B
DeepSeek
1.3BCodingQ8_01.8GB
Granite 3.0 1B-A400M
IBM
1.3BChat / LLMQ4_K_M1.3GB
Llama 3.2 1B Instruct
Meta
1.24BChat / LLMQ8_01.7GB
TinyLlama 1.1B
TinyLlama
1.1BChat / LLMQ8_01.6GB
Gemma 3 1B
1BChat / LLMQ8_01.5GB
Falcon 3 1B
TII
1BChat / LLMQ8_02.2GB
Stable Diffusion 2.1 Base (CoreML)
Stability AI / Apple
0.86BImage GenerationCoreML-Palettized1.6GB
Stable Diffusion 1.5 (CoreML)
Runway
0.86BImage GenerationCoreML-Palettized2.5GB
Stable Diffusion 1.5 (GGUF)
Runway / GPUStack
0.86BImage GenerationQ8_02.3GB
Stable Diffusion 2.1 (GGUF)
Stability AI
0.86BImage GenerationQ8_02.7GB
Whisper Large v3 Turbo
OpenAI
0.81BSpeech RecognitionQ8_02.0GB
Whisper Medium
OpenAI
0.77BSpeech RecognitionQ8_01.9GB
Distil-Whisper Large v3
HuggingFace
0.76BSpeech RecognitionQ8_01.9GB
BGE Reranker v2 M3
BAAI
0.568BRerankerFP161.6GB
Qwen 2.5 0.5B
Alibaba
0.5BChat / LLMQ8_01.1GB
Qwen 2.5 Coder 0.5B
Alibaba
0.5BCodingQ8_01.1GB
Danube 3 500M
H2O.ai
0.5BChat / LLMQ8_01.0GB
SmolLM2 360M
HuggingFace
0.36BChat / LLMQ8_00.9GB
BGE Large EN v1.5
BAAI
0.335BEmbeddingFP161.1GB
MusicGen Small
Meta
0.3BAudio GenerationONNX-Q4F160.8GB
Whisper Small
OpenAI
0.24BSpeech RecognitionQ8_00.9GB
Nomic Embed Text v1.5
Nomic AI
0.137BEmbeddingFP160.8GB
SmolLM2 135M
HuggingFace
0.135BChat / LLMFP160.8GB
Kokoro 82M TTS
Kokoro
0.082BText to SpeechONNX-Q8F160.6GB
Whisper Base
OpenAI
0.074BSpeech RecognitionQ8_00.3GB
Whisper Base English
OpenAI
0.074BSpeech RecognitionQ8_00.3GB
Whisper Tiny English (Quantized)
OpenAI
0.039BSpeech RecognitionQ5_10.1GB
Whisper Tiny
OpenAI
0.039BSpeech RecognitionQ8_00.2GB
BGE Small EN v1.5
BAAI
0.033BEmbeddingQ8_00.1GB
Snowflake Arctic Embed S
Snowflake
0.033BEmbeddingQ8_00.1GB
Jina Reranker Tiny EN
Jina AI
0.033BRerankerFP160.1GB
all-MiniLM-L6-v2
Sentence Transformers
0.023BEmbeddingQ8_00.1GB
Piper TTS - Amy (English)
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - Lessac (English)
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - LibriTTS-R (English)
Rhasspy
0.02BText to SpeechONNX0.6GB
Piper TTS - Spanish (MLS)
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - French (Siwis)
Rhasspy
0.02BText to SpeechONNX0.5GB
Piper TTS - German (Thorsten)
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - Chinese (Huayan)
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - Japanese (Kokoro)
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - Korean
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - Russian (Irina)
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - Portuguese (Faber)
Rhasspy
0.02BText to SpeechONNX0.1GB
Piper TTS - Italian (Riccardo)
Rhasspy
0.02BText to SpeechONNX0.5GB
Piper TTS - Arabic (Kareem)
Rhasspy
0.02BText to SpeechONNX0.1GB
A
Great (6 models)
Runs well with good performance
B
Good (3 models)
Runs but may be tight on memory
C
Possible (13 models)
Needs partial CPU offloading, slower performance
EXAONE 3.5 7.8B
LG AI
7.8BChat / LLMQ4_K_M4.9GB
InternLM 2.5 7B
Shanghai AI Lab
7.7BChat / LLMQ4_K_M4.9GB
Qwen 2.5 Coder 7B
Alibaba
7.6BCodingQ4_K_M4.9GB
Mistral 7B Instruct v0.3
Mistral AI
7.3BChat / LLMQ4_K_M4.6GB
LLaVA 1.6 7B
LLaVA
7BMultimodalQ4_K_M5.0GB
Falcon 3 7B
TII
7BChat / LLMQ4_K_M5.0GB
OLMo 2 7B
Allen AI
7BChat / LLMQ4_K_M4.7GB
OpenChat 3.5 7B
OpenChat
7BChat / LLMQ4_K_M4.6GB
StarCoder2 7B
BigCode
7BCodingQ4_K_M4.7GB
Code Llama 7B
Meta
7BCodingQ4_K_M4.3GB
OLMoE 1B-7B
AI2
6.9BChat / LLMQ4_K_M4.4GB
DeepSeek Coder 6.7B
DeepSeek
6.7BCodingQ4_K_M4.3GB
SDXL Turbo (GGUF)
Stability AI
3.5BImage GenerationQ5_05.0GB
D
Struggling (23 models)
Heavy offloading required, very slow
Code Llama 13B Instruct
Meta
13BCodingQ4_K_M7.8GB
Gemma 3 12B
12BChat / LLMQ4_K_M7.3GB
Mistral Nemo 12B
Mistral AI
12BChat / LLMQ4_K_M7.5GB
Magnum v4 12B
Anthracite
12BChat / LLMQ4_K_M7.5GB
Rocinante 12B v1.1
TheDrummer
12BChat / LLMQ4_K_M7.5GB
Mistral Nemo Base 12B
Mistral AI
12BChat / LLMQ4_K_M7.7GB
Solar 10.7B
Upstage
10.7BChat / LLMQ4_K_M6.5GB
Falcon 3 10B
TII
10BChat / LLMQ4_K_M6.4GB
Gemma 2 9B Instruct
9.2BChat / LLMQ5_K_M6.7GB
Yi 1.5 9B Chat
01.AI
9BChat / LLMQ4_K_M5.5GB
Yi Coder 9B
01.AI
9BCodingQ4_K_M5.5GB
Gemma 3 MoE 9B
9BChat / LLMQ4_K_M7.0GB
CodeGemma 7B
8.5BCodingQ4_K_M5.5GB
DeepSeek R1 Distill 8B
DeepSeek
8BChat / LLMQ5_K_M5.8GB
Llama 3.1 8B Instruct
Meta
8BChat / LLMQ5_K_M5.8GB
Granite 3.3 8B
IBM
8BChat / LLMQ4_K_M5.1GB
Dolphin 3.0 Llama 3.1 8B
Cognitive Computations
8BChat / LLMQ4_K_M5.1GB
NeuralDaredevil 8B (abliterated)
mlabonne
8BChat / LLMQ4_K_M5.1GB
Llama 3.1 8B Instruct (abliterated)
mlabonne
8BChat / LLMQ4_K_M5.1GB
Stheno L3 8B v3.2
Sao10K
8BChat / LLMQ4_K_M5.1GB
Qwen3 8B Base
Alibaba
8BChat / LLMQ4_K_M5.3GB
Qwen 2.5 7B Instruct
Alibaba
7.6BChat / LLMQ5_K_M6.2GB
Stable Audio Open
Stability AI
1BAudio GenerationFP166.0GB
F
Cannot Run (33 models)
Insufficient hardware to run this model
Qwen3 235B-A22B
Alibaba
235BChat / LLMQ4_K_M144.0GB
Mixtral 8x22B Instruct
Mistral AI
141BChat / LLMQ4_K_M88.0GB
Magnum v4 72B
Anthracite
72BChat / LLMBF16144.5GB
Llama 3.1 70B Instruct
Meta
70BChat / LLMFP16142.0GB
Euryale L3.3 70B v2.3
Sao10K
70BChat / LLMBF16140.5GB
Llama 3.1 70B (lorablated)
mlabonne
70BChat / LLMBF16140.5GB
Mixtral 8x7B Instruct
Mistral AI
46.7BChat / LLMQ5_K_M30.5GB
Phi-3.5 MoE
Microsoft
41.9BChat / LLMQ4_K_M24.1GB
Qwen 2.5 32B
Alibaba
32BChat / LLMQ4_K_M19.0GB
Skyfall 31B v4.2
TheDrummer
31BChat / LLMBF1662.5GB
Qwen3 30B-A3B
Alibaba
30.5BChat / LLMQ8_036.0GB
Gemma 3 27B
27BChat / LLMQ4_K_M15.9GB
Dolphin Mistral 24B (Venice Edition)
Cognitive Computations
24BChat / LLMBF1648.5GB
Dolphin 3.0 R1 Mistral 24B
Cognitive Computations
24BChat / LLMBF1648.5GB
Cydonia 24B v4.3
TheDrummer
24BChat / LLMBF1648.5GB
Mistral Small 22B
Mistral AI
22BChat / LLMQ4_K_M12.9GB
Codestral 22B (abliterated)
failspy
22BCodingBF1644.5GB
Magnum v4 22B
Anthracite
22BChat / LLMBF1644.5GB
DeepSeek MoE 16B
DeepSeek
16.4BChat / LLMQ4_K_M11.0GB
Rocinante XL 16B v1
TheDrummer
16BChat / LLMBF1632.5GB
Phi-4
Microsoft
14BChat / LLMQ8_015.0GB
Qwen 2.5 14B
Alibaba
14BChat / LLMQ8_015.1GB
Qwen 2.5 Coder 14B
Alibaba
14BCodingQ8_015.1GB
HunyuanVideo 13B
Tencent
13Bvideo-genFP1660.0GB
FLUX.1 Schnell (GGUF)
Black Forest Labs
12BImage GenerationQ5_014.0GB
FLUX.1 Dev (GGUF)
Black Forest Labs
12BImage GenerationQ5_014.0GB
Mochi 1 Preview
Genmo
10Bvideo-genFP1630.0GB
Wan 2.2 TI2V 5B
Wan-AI
5Bvideo-genFP1616.0GB
CogVideoX 5B
THUDM
5Bvideo-genFP1616.0GB
Stable Diffusion 3 Medium (GGUF)
Stability AI
2.5BImage GenerationQ8_09.2GB
Hunyuan3D 2
Tencent
2.5B3d-genFP1616.0GB
ACE-Step 1.5XL
ACE Studio
1.5BAudio GenerationFP168.0GB
TRELLIS Image Large
Microsoft
1.2B3d-genFP1612.0GB
Similar GPUs
NVIDIA GeForce GTX 1050 Ti
4GB$139nvidia
NVIDIA GeForce GTX 1650 SUPER
4GB$159nvidia
NVIDIA GeForce GTX 1060 6GB
6GB$249nvidia
NVIDIA GeForce GTX 1660
6GB$219nvidia
NVIDIA GeForce GTX 1660 SUPER
6GB$229nvidia
NVIDIA GeForce GTX 1660 Ti
6GB$279nvidia
NVIDIA GeForce RTX 2060
6GB$349nvidia
NVIDIA RTX 3060 Laptop
6GBnvidia
AMD Radeon RX 6600
8GB$329amd
AMD Radeon RX 6600 XT
8GB$379amd
AMD Radeon RX 7600
8GB$269amd
Apple Silicon (Unknown)
8GBapple
Frequently Asked Questions
What AI models can I run on the NVIDIA GeForce GTX 1650?
The NVIDIA GeForce GTX 1650 with 4GB VRAM can run 76 AI models smoothly (grade B or better), including Granite 3.0 3B-A800M, Llama 3.2 3B Instruct, Qwen 2.5 3B and 73 more.
How much VRAM does the NVIDIA GeForce GTX 1650 have?
The NVIDIA GeForce GTX 1650 has 4GB of VRAM, which is sufficient for running many local AI models.
Is the NVIDIA GeForce GTX 1650 good for AI?
With 4GB VRAM, the NVIDIA GeForce GTX 1650 can run 76 out of 145 models in our database at grade B or better. You may need smaller quantizations for larger models.