LLM Finder

Match local LLMs to your GPU. Filter by VRAM + use case, then copy the download command.

GPU VRAM

8 GB 12 GB 16 GB 24 GB 48 GB 80 GB

Use Case

all coding chat research creative math

22 models fit 24GB for research

Data powered by llmfit ↗

Qwen2.5-Coder-32B-Instruct

32Bq4_K_M • 22.3GB

Family: Qwen

Estimated speed: 19-29 tok/s

Context: 33k

License: Apache-2.0

codingchatmathresearch

ollama pull qwen2.5-coder:32b

Yi-1.5-34B-Chat

34Bq4_K_M • 23.6GB

Family: Yi

Estimated speed: 18-28 tok/s

Context: 33k

License: Apache-2.0

chatresearchcreativemath

ollama pull yi:34b

Qwen2.5-32B-Instruct

32Bq4_K_M • 22.3GB

Family: Qwen

Estimated speed: 20-30 tok/s

Context: 33k

License: Apache-2.0

chatcodingresearchmathcreative

ollama pull qwen2.5:32b

Phi-4-14B-Instruct

14Bq8_0 • 18.3GB

Family: Phi

Estimated speed: 31-41 tok/s

Context: 128k

License: MIT

codingmathresearchchat

ollama pull phi4:14b

Phi-3-medium-128k-instruct

14Bq8_0 • 18.3GB

Family: Phi

Estimated speed: 32-42 tok/s

Context: 131k

License: MIT

codingchatresearchmath

ollama pull phi3:medium

DeepSeek-R1-Distill-Llama-8B

8Bfp16 • 20.1GB

Family: DeepSeek

Estimated speed: 28-38 tok/s

Context: 33k

License: MIT

mathresearchchat

ollama pull deepseek-r1:8b

Qwen2.5-14B-Instruct

14Bq8_0 • 18.3GB

Family: Qwen

Estimated speed: 33-43 tok/s

Context: 33k

License: Apache-2.0

chatcodingresearchmath

ollama pull qwen2.5:14b

Gemma-2-27B-Instruct

27Bq5_K_M • 23.5GB

Family: Gemma

Estimated speed: 22-32 tok/s

Context: 8k

License: Gemma Terms

chatcodingresearchcreativemath

ollama pull gemma2:27b

DeepSeek-R1-Distill-Qwen-14B

14Bq8_0 • 18.3GB

Family: DeepSeek

Estimated speed: 34-44 tok/s

Context: 33k

License: MIT

mathresearchcoding

ollama pull deepseek-r1:14b

Ministral-8B-Instruct

8Bfp16 • 20.1GB

Family: Mistral

Estimated speed: 31-41 tok/s

Context: 128k

License: Mistral Research

chatresearch

ollama pull ministral:8b

Command-R-7B

7Bfp16 • 17.9GB

Family: Cohere

Estimated speed: 36-46 tok/s

Context: 128k

License: CC-BY-NC

researchchatcreative

ollama pull command-r:7b

Mixtral-8x7B-Instruct

46.7B MoEq2_K • 19.6GB

Family: Mistral

Estimated speed: 32-42 tok/s

Context: 33k

License: Apache-2.0

chatcodingresearchcreative

ollama pull mixtral:8x7b

Mistral-7B-Instruct-v0.3

7Bfp16 • 17.9GB

Family: Mistral

Estimated speed: 37-47 tok/s

Context: 33k

License: Apache-2.0

chatresearch

ollama pull mistral:7b

Yi-1.5-9B-Chat

9Bfp16 • 22.3GB

Family: Yi

Estimated speed: 26-36 tok/s

Context: 33k

License: Apache-2.0

chatresearchcreative

ollama pull yi:9b

Qwen2.5-7B-Instruct

7Bfp16 • 17.9GB

Family: Qwen

Estimated speed: 37-47 tok/s

Context: 33k

License: Apache-2.0

chatcodingresearchmath

ollama pull qwen2.5:7b

Llama-3.1-8B-Instruct

8Bfp16 • 20.1GB

Family: Llama

Estimated speed: 33-43 tok/s

Context: 131k

License: Llama 3.1 Community

chatresearchcreative

ollama pull llama3.1:8b

Gemma-2-9B-Instruct

9Bfp16 • 22.3GB

Family: Gemma

Estimated speed: 28-38 tok/s

Context: 8k

License: Gemma Terms

chatresearchcreative

ollama pull gemma2:9b

Neural-Chat-7B-v3.3

7Bfp16 • 17.9GB

Family: Intel

Estimated speed: 39-49 tok/s

Context: 33k

License: Apache-2.0

chatresearch

ollama pull neural-chat

Phi-4-mini-instruct

3.8Bfp16 • 10.9GB

Family: Phi

Estimated speed: 56-66 tok/s

Context: 128k

License: MIT

chatcodingresearch

ollama pull phi4-mini

Command-R-35B

35Bq3_K_M • 19.2GB

Family: Cohere

Estimated speed: 36-46 tok/s

Context: 128k

License: CC-BY-NC

researchchatcoding

ollama pull command-r:35b

DeepSeek-R1-Distill-Qwen-7B

7Bfp16 • 17.9GB

Family: DeepSeek

Estimated speed: 40-50 tok/s

Context: 33k

License: MIT

mathresearchcoding

ollama pull deepseek-r1:7b

Qwen2.5-3B-Instruct

3Bfp16 • 9.1GB

Family: Qwen

Estimated speed: 65-75 tok/s

Context: 33k

License: Apache-2.0

chatcodingresearch

ollama pull qwen2.5:3b