Tokenstead - Find the right AI model for your hardware

M3 Pro 18GB

37 models / sorted by best_quality

Quality Fastest Least mem Newest

Model Fit Speed Memory Scores / Run

Runs comfortably

Qwen3.8-27B UD-IQ1_S Unsloth

6.2GB 18k cap

Runs comfortably

11 tok/s

ⓘ estimated

6 / 18GB

88 / 89 / 85

FIT Runs comfortably

SPEED 11 tok/s

ⓘ estimated

MEMORY 6/18GB

Qwen3.8-27B UD-Q2_K_XL Unsloth

9.8GB 12k cap

Runs comfortably

7 tok/s

ⓘ estimated

10 / 18GB

88 / 89 / 85

FIT Runs comfortably

SPEED 7 tok/s

ⓘ estimated

MEMORY 10/18GB

Gemma 4 26B A4B NVFP4

14.0GB 4k cap

Runs comfortably

18 tok/s

ⓘ estimated

14 / 18GB

85 / 88 / 84

FIT Runs comfortably

SPEED 18 tok/s

ⓘ estimated

MEMORY 14/18GB

Gemma 4 26B A4B Q4_K_M

13.0GB 2k cap

Runs comfortably

19 tok/s

ⓘ estimated

13 / 18GB

85 / 88 / 84

FIT Runs comfortably

SPEED 19 tok/s

ⓘ estimated

MEMORY 13/18GB

Ornith-1.5-9B Q8_0

9.8GB 16k cap

Runs comfortably

8 tok/s

ⓘ estimated

10 / 18GB

78 / 76 / 74

FIT Runs comfortably

SPEED 8 tok/s

ⓘ estimated

MEMORY 10/18GB

Ornith-1.5-9B Q6_K

7.6GB 20k cap

Runs comfortably

9 tok/s

ⓘ estimated

8 / 18GB

78 / 76 / 74

FIT Runs comfortably

SPEED 9 tok/s

ⓘ estimated

MEMORY 8/18GB

Ornith-1.5-9B Q5_K_M

6.6GB 22k cap

Runs comfortably

10 tok/s

ⓘ estimated

7 / 18GB

78 / 76 / 74

FIT Runs comfortably

SPEED 10 tok/s

ⓘ estimated

MEMORY 7/18GB

Ornith-1.5-9B Q4_K_M

5.8GB 24k cap

Runs comfortably

12 tok/s

ⓘ estimated

6 / 18GB

78 / 76 / 74

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 6/18GB

Qwen3 30B A3B Q4_K_M

15.5GB 4k cap

Runs comfortably

24 tok/s

ⓘ estimated

16 / 18GB

75 / 78 / 72

FIT Runs comfortably

SPEED 24 tok/s

ⓘ estimated

MEMORY 16/18GB

Phi-4 14B Q4_K_M

8.5GB 16k ctx

Runs comfortably

9 tok/s

ⓘ estimated

9 / 18GB

75 / 78 / 70

FIT Runs comfortably

SPEED 9 tok/s

ⓘ estimated

MEMORY 9/18GB

Phi-4 14B Q8_0

15.0GB 7k cap

Runs comfortably

5 tok/s

ⓘ estimated

15 / 18GB

75 / 78 / 70

FIT Runs comfortably

SPEED 5 tok/s

ⓘ estimated

MEMORY 15/18GB

Qwen3 14B Q4_K_M

9.0GB 18k cap

Runs comfortably

8 tok/s

ⓘ estimated

9 / 18GB

72 / 75 / 70

FIT Runs comfortably

SPEED 8 tok/s

ⓘ estimated

MEMORY 9/18GB

Qwen3 14B Q8_0

15.0GB 6k cap

Runs comfortably

5 tok/s

ⓘ estimated

15 / 18GB

72 / 75 / 70

FIT Runs comfortably

SPEED 5 tok/s

ⓘ estimated

MEMORY 15/18GB

Gemma 4 12B Q8_0

13.0GB 11k cap

Runs comfortably

6 tok/s

ⓘ estimated

13 / 18GB

70 / 74 / 72

FIT Runs comfortably

SPEED 6 tok/s

ⓘ estimated

MEMORY 13/18GB

Gemma 4 12B Q4_K_M

7.2GB 23k cap

Runs comfortably

10 tok/s

ⓘ estimated

7 / 18GB

70 / 74 / 72

FIT Runs comfortably

SPEED 10 tok/s

ⓘ estimated

MEMORY 7/18GB

Qwen3 8B Q4_K_M

5.2GB 25k cap

Runs comfortably

12 tok/s

ⓘ estimated

5 / 18GB

67 / 70 / 65

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 5/18GB

Qwen3 8B Q8_0

8.5GB 19k cap

Runs comfortably

9 tok/s

ⓘ estimated

9 / 18GB

67 / 70 / 65

FIT Runs comfortably

SPEED 9 tok/s

ⓘ estimated

MEMORY 9/18GB

Gemma 3 12B Q8_0

13.5GB 10k cap

Runs comfortably

6 tok/s

ⓘ estimated

14 / 18GB

62 / 65 / 68

FIT Runs comfortably

SPEED 6 tok/s

ⓘ estimated

MEMORY 14/18GB

Gemma 3 12B Q4_K_M

7.5GB 25k cap

Runs comfortably

10 tok/s

ⓘ estimated

8 / 18GB

62 / 65 / 68

FIT Runs comfortably

SPEED 10 tok/s

ⓘ estimated

MEMORY 8/18GB

Mistral Nemo 12B Q4_K_M

7.3GB 26k cap

Runs comfortably

10 tok/s

ⓘ estimated

7 / 18GB

58 / 60 / 62

FIT Runs comfortably

SPEED 10 tok/s

ⓘ estimated

MEMORY 7/18GB

Mistral Nemo 12B Q8_0

13.2GB 11k cap

Runs comfortably

6 tok/s

ⓘ estimated

13 / 18GB

58 / 60 / 62

FIT Runs comfortably

SPEED 6 tok/s

ⓘ estimated

MEMORY 13/18GB

Qwen3 4B Q4_K_M

2.6GB 32k ctx

Runs comfortably

24 tok/s

ⓘ estimated

3 / 18GB

58 / 60 / 56

FIT Runs comfortably

SPEED 24 tok/s

ⓘ estimated

MEMORY 3/18GB

Phi-4 Mini Q8_0

4.1GB 69k cap

Runs comfortably

18 tok/s

ⓘ estimated

4 / 18GB

55 / 58 / 52

FIT Runs comfortably

SPEED 18 tok/s

ⓘ estimated

MEMORY 4/18GB

Phi-4 Mini Q4_K_M

2.5GB 77k cap

Runs comfortably

27 tok/s

ⓘ estimated

3 / 18GB

55 / 58 / 52

FIT Runs comfortably

SPEED 27 tok/s

ⓘ estimated

MEMORY 3/18GB

Llama 3.1 8B Instruct Q8_0

8.0GB 20k cap

Runs comfortably

9 tok/s

ⓘ estimated

8 / 18GB

52 / 55 / 62

FIT Runs comfortably

SPEED 9 tok/s

ⓘ estimated

MEMORY 8/18GB

Llama 3.1 8B Instruct FP16

16.0GB 4k cap

Runs comfortably

5 tok/s

ⓘ estimated

16 / 18GB

52 / 55 / 62

FIT Runs comfortably

SPEED 5 tok/s

ⓘ estimated

MEMORY 16/18GB

Llama 3.1 8B Instruct Q4_K_M

4.5GB 27k cap

Runs comfortably

14 tok/s

ⓘ estimated

5 / 18GB

52 / 55 / 62

FIT Runs comfortably

SPEED 14 tok/s

ⓘ estimated

MEMORY 5/18GB

Gemma 3 4B Q8_0

5.0GB 43k cap

Runs comfortably

14 tok/s

ⓘ estimated

5 / 18GB

42 / 45 / 55

FIT Runs comfortably

SPEED 14 tok/s

ⓘ estimated

MEMORY 5/18GB

Gemma 3 4B Q4_K_M

3.0GB 50k cap

Runs comfortably

21 tok/s

ⓘ estimated

3 / 18GB

42 / 45 / 55

FIT Runs comfortably

SPEED 21 tok/s

ⓘ estimated

MEMORY 3/18GB

Llama 3.2 3B Instruct Q8_0

3.5GB 60k cap

Runs comfortably

20 tok/s

ⓘ estimated

4 / 18GB

35 / 38 / 50

FIT Runs comfortably

SPEED 20 tok/s

ⓘ estimated

MEMORY 4/18GB

Llama 3.2 3B Instruct Q4_K_M

2.0GB 66k cap

Runs comfortably

30 tok/s

ⓘ estimated

2 / 18GB

35 / 38 / 50

FIT Runs comfortably

SPEED 30 tok/s

ⓘ estimated

MEMORY 2/18GB

Mellum2.1 12B Thinking Q4_K_M

7.0GB 131k ctx

Runs comfortably

54 tok/s

ⓘ estimated

7 / 18GB

FIT Runs comfortably

SPEED 54 tok/s

ⓘ estimated

MEMORY 7/18GB

Runs tight

Pokee-Isaac 28B Q4_K_M

16.0GB ctx headroom ~0

Runs tight

5 tok/s

ⓘ estimated

16 / 18GB

90 / 88 / 88

FIT Runs tight

SPEED 5 tok/s

ⓘ estimated

MEMORY 16/18GB

Qwen3.8-27B NVFP4 Unsloth

15.2GB ctx headroom ~0

Runs tight

5 tok/s

ⓘ estimated

15 / 18GB

88 / 89 / 85

FIT Runs tight

SPEED 5 tok/s

ⓘ estimated

MEMORY 15/18GB

Qwen3.6 27B Q4_K_M

16.5GB 2k cap

Runs tight

5 tok/s

ⓘ estimated

17 / 18GB

85 / 87 / 82

FIT Runs tight

SPEED 5 tok/s

ⓘ estimated

MEMORY 17/18GB

Gemma 3 27B Q4_K_M

17.0GB 1k cap

Runs tight

4 tok/s

ⓘ estimated

17 / 18GB

68 / 71 / 72

FIT Runs tight

SPEED 4 tok/s

ⓘ estimated

MEMORY 17/18GB

LFM2.5-24B-A2B Q4_K_M

13.7GB 25k cap

Runs tight

62 tok/s

ⓘ estimated

14 / 18GB

FIT Runs tight

SPEED 62 tok/s

ⓘ estimated

MEMORY 14/18GB

Coding alternatives

Hosted coding assistants for comparison - per seat

Cursor Hobby

AI-first code editor. Uses frontier models (GPT-4, Claude). $20/user Business...

$0.0

/MONTH

GitHub Copilot Individual

AI coding assistant by GitHub. GPT-4o and Claude-based completions. $19/seat ...

$10.0

/MONTH

GitHub Copilot Business

AI coding assistant by GitHub. GPT-4o and Claude-based completions. $19/seat ...

$19.0

/MONTH

Cursor Pro

AI-first code editor. Uses frontier models (GPT-4, Claude). $20/user Business...

$20.0

/MONTH

GitHub Copilot Enterprise

AI coding assistant by GitHub. GPT-4o and Claude-based completions. $19/seat ...

$39.0

/MONTH

Cursor Business

AI-first code editor. Uses frontier models (GPT-4, Claude). $20/user Business...

$40.0

/MONTH

Per-token monthly figures are estimates at typical usage. Prices verified periodically - always confirm current pricing on the provider's website.

On a monthly budget? Compare owning hardware vs renting cloud →

Tokenstead community

Not sure this is the right fit?

Post your rig and the shortlist in the community and get a sanity check from people running the same memory and stack.

Join the community Share your rig