Tokenstead - Find the right AI model for your hardware

M4 Max Binned 36GB

67 models / sorted by best_quality

Quality Fastest Least mem Newest

Model Fit Speed Memory Scores / Run

Runs comfortably

Ornith-1.5-35B-A3B Q6_K

29.2GB 10k cap

Runs comfortably

53 tok/s

ⓘ estimated

29 / 36GB

91 / 89 / 86

FIT Runs comfortably

SPEED 53 tok/s

ⓘ estimated

MEMORY 29/36GB

Ornith-1.5-35B-A3B Q5_K_M

25.3GB 18k cap

Runs comfortably

56 tok/s

ⓘ estimated

25 / 36GB

91 / 89 / 86

FIT Runs comfortably

SPEED 56 tok/s

ⓘ estimated

MEMORY 25/36GB

Ornith-1.5-35B-A3B Q4_K_M

21.7GB 25k cap

Runs comfortably

61 tok/s

ⓘ estimated

22 / 36GB

91 / 89 / 86

FIT Runs comfortably

SPEED 61 tok/s

ⓘ estimated

MEMORY 22/36GB

Qwen3.6 35B A3B Q4_K_M

20.0GB 29k cap

Runs comfortably

62 tok/s

ⓘ estimated

20 / 36GB

90 / 91 / 87

FIT Runs comfortably

SPEED 62 tok/s

ⓘ estimated

MEMORY 20/36GB

Pokee-Isaac 28B Q4_K_M

16.0GB 30k cap

Runs comfortably

13 tok/s

ⓘ estimated

16 / 36GB

90 / 88 / 88

FIT Runs comfortably

SPEED 13 tok/s

ⓘ estimated

MEMORY 16/36GB

Pokee-Isaac 28B Q8_0

28.0GB 10k cap

Runs comfortably

8 tok/s

ⓘ estimated

28 / 36GB

90 / 88 / 88

FIT Runs comfortably

SPEED 8 tok/s

ⓘ estimated

MEMORY 28/36GB

Qwen3.8-27B UD-Q4_K_XL Unsloth

17.6GB 30k cap

Runs comfortably

12 tok/s

ⓘ estimated

18 / 36GB

88 / 89 / 85

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 18/36GB

Qwen3.8-27B UD-Q2_K_XL Unsloth

9.8GB 45k cap

Runs comfortably

20 tok/s

ⓘ estimated

10 / 36GB

88 / 89 / 85

FIT Runs comfortably

SPEED 20 tok/s

ⓘ estimated

MEMORY 10/36GB

Qwen3.8-27B UD-IQ1_S Unsloth

6.2GB 50k cap

Runs comfortably

29 tok/s

ⓘ estimated

6 / 36GB

88 / 89 / 85

FIT Runs comfortably

SPEED 29 tok/s

ⓘ estimated

MEMORY 6/36GB

Qwen3.8-27B NVFP4 Unsloth

15.2GB 32k cap

Runs comfortably

14 tok/s

ⓘ estimated

15 / 36GB

88 / 89 / 85

FIT Runs comfortably

SPEED 14 tok/s

ⓘ estimated

MEMORY 15/36GB

Gemma 4 31B QAT

18.0GB 21k cap

Runs comfortably

12 tok/s

ⓘ estimated

18 / 36GB

86 / 89 / 85

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 18/36GB

Gemma 4 31B NVFP4

16.0GB 20k cap

Runs comfortably

13 tok/s

ⓘ estimated

16 / 36GB

86 / 89 / 85

FIT Runs comfortably

SPEED 13 tok/s

ⓘ estimated

MEMORY 16/36GB

Gemma 4 31B Q4_K_M

18.3GB 20k cap

Runs comfortably

11 tok/s

ⓘ estimated

18 / 36GB

86 / 89 / 85

FIT Runs comfortably

SPEED 11 tok/s

ⓘ estimated

MEMORY 18/36GB

Gemma 4 26B A4B Q8_0

25.0GB 12k cap

Runs comfortably

37 tok/s

ⓘ estimated

25 / 36GB

85 / 88 / 84

FIT Runs comfortably

SPEED 37 tok/s

ⓘ estimated

MEMORY 25/36GB

Gemma 4 26B A4B Q4_K_M

13.0GB 28k cap

Runs comfortably

51 tok/s

ⓘ estimated

13 / 36GB

85 / 88 / 84

FIT Runs comfortably

SPEED 51 tok/s

ⓘ estimated

MEMORY 13/36GB

Gemma 4 26B A4B NVFP4

14.0GB 29k cap

Runs comfortably

49 tok/s

ⓘ estimated

14 / 36GB

85 / 88 / 84

FIT Runs comfortably

SPEED 49 tok/s

ⓘ estimated

MEMORY 14/36GB

Qwen3.6 27B Q4_K_M

16.5GB 27k cap

Runs comfortably

13 tok/s

ⓘ estimated

17 / 36GB

85 / 87 / 82

FIT Runs comfortably

SPEED 13 tok/s

ⓘ estimated

MEMORY 17/36GB

Qwen3.6 27B Q8_0

28.0GB 11k cap

Runs comfortably

8 tok/s

ⓘ estimated

28 / 36GB

85 / 87 / 82

FIT Runs comfortably

SPEED 8 tok/s

ⓘ estimated

MEMORY 28/36GB

Qwen3 32B Q6_K

28.5GB 10k cap

Runs comfortably

8 tok/s

ⓘ estimated

29 / 36GB

80 / 82 / 78

FIT Runs comfortably

SPEED 8 tok/s

ⓘ estimated

MEMORY 29/36GB

Qwen3 32B Q5_K_M

24.5GB 16k cap

Runs comfortably

9 tok/s

ⓘ estimated

25 / 36GB

80 / 82 / 78

FIT Runs comfortably

SPEED 9 tok/s

ⓘ estimated

MEMORY 25/36GB

Qwen3 32B Q4_K_M

20.0GB 22k cap

Runs comfortably

11 tok/s

ⓘ estimated

20 / 36GB

80 / 82 / 78

FIT Runs comfortably

SPEED 11 tok/s

ⓘ estimated

MEMORY 20/36GB

Ornith-1.5-9B Q5_K_M

6.6GB 58k cap

Runs comfortably

28 tok/s

ⓘ estimated

7 / 36GB

78 / 76 / 74

FIT Runs comfortably

SPEED 28 tok/s

ⓘ estimated

MEMORY 7/36GB

Ornith-1.5-9B Q6_K

7.6GB 56k cap

Runs comfortably

26 tok/s

ⓘ estimated

8 / 36GB

78 / 76 / 74

FIT Runs comfortably

SPEED 26 tok/s

ⓘ estimated

MEMORY 8/36GB

Ornith-1.5-9B Q8_0

9.8GB 52k cap

Runs comfortably

21 tok/s

ⓘ estimated

10 / 36GB

78 / 76 / 74

FIT Runs comfortably

SPEED 21 tok/s

ⓘ estimated

MEMORY 10/36GB

Ornith-1.5-9B BF16

18.4GB 34k cap

Runs comfortably

12 tok/s

ⓘ estimated

18 / 36GB

78 / 76 / 74

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 18/36GB

Ornith-1.5-9B Q4_K_M

5.8GB 60k cap

Runs comfortably

31 tok/s

ⓘ estimated

6 / 36GB

78 / 76 / 74

FIT Runs comfortably

SPEED 31 tok/s

ⓘ estimated

MEMORY 6/36GB

Qwen3 30B A3B Q4_K_M

15.5GB 37k cap

Runs comfortably

65 tok/s

ⓘ estimated

16 / 36GB

75 / 78 / 72

FIT Runs comfortably

SPEED 65 tok/s

ⓘ estimated

MEMORY 16/36GB

Qwen3 30B A3B Q8_0

31.0GB 9k cap

Runs comfortably

46 tok/s

ⓘ estimated

31 / 36GB

75 / 78 / 72

FIT Runs comfortably

SPEED 46 tok/s

ⓘ estimated

MEMORY 31/36GB

Phi-4 14B Q4_K_M

8.5GB 16k ctx

Runs comfortably

24 tok/s

ⓘ estimated

9 / 36GB

75 / 78 / 70

FIT Runs comfortably

SPEED 24 tok/s

ⓘ estimated

MEMORY 9/36GB

Phi-4 14B FP16

29.0GB 16k ctx

Runs comfortably

8 tok/s

ⓘ estimated

29 / 36GB

75 / 78 / 70

FIT Runs comfortably

SPEED 8 tok/s

ⓘ estimated

MEMORY 29/36GB

Phi-4 14B Q8_0

15.0GB 16k ctx

Runs comfortably

15 tok/s

ⓘ estimated

15 / 36GB

75 / 78 / 70

FIT Runs comfortably

SPEED 15 tok/s

ⓘ estimated

MEMORY 15/36GB

Xing4.0-29B-A4B Q4_K_M

20.1GB 17k cap

Runs comfortably

32 tok/s

ⓘ estimated

20 / 36GB

75 / 73 / 72

FIT Runs comfortably

SPEED 32 tok/s

ⓘ estimated

MEMORY 20/36GB

Llama 3.3 70B Instruct Q2_K

21.5GB 9k cap

Runs comfortably

9 tok/s

ⓘ estimated

22 / 36GB

72 / 77 / 80

FIT Runs comfortably

SPEED 9 tok/s

ⓘ estimated

MEMORY 22/36GB

Llama 3.3 70B Instruct Q3_K_M

30.0GB 4k cap

Runs comfortably

7 tok/s

ⓘ estimated

30 / 36GB

72 / 77 / 80

FIT Runs comfortably

SPEED 7 tok/s

ⓘ estimated

MEMORY 30/36GB

Qwen3 14B Q8_0

15.0GB 42k cap

Runs comfortably

14 tok/s

ⓘ estimated

15 / 36GB

72 / 75 / 70

FIT Runs comfortably

SPEED 14 tok/s

ⓘ estimated

MEMORY 15/36GB

Qwen3 14B Q4_K_M

9.0GB 54k cap

Runs comfortably

22 tok/s

ⓘ estimated

9 / 36GB

72 / 75 / 70

FIT Runs comfortably

SPEED 22 tok/s

ⓘ estimated

MEMORY 9/36GB

Gemma 4 12B Q8_0

13.0GB 51k cap

Runs comfortably

17 tok/s

ⓘ estimated

13 / 36GB

70 / 74 / 72

FIT Runs comfortably

SPEED 17 tok/s

ⓘ estimated

MEMORY 13/36GB

Gemma 4 12B Q4_K_M

7.2GB 63k cap

Runs comfortably

27 tok/s

ⓘ estimated

7 / 36GB

70 / 74 / 72

FIT Runs comfortably

SPEED 27 tok/s

ⓘ estimated

MEMORY 7/36GB

Gemma 3 27B Q8_0

28.0GB 10k cap

Runs comfortably

8 tok/s

ⓘ estimated

28 / 36GB

68 / 71 / 72

FIT Runs comfortably

SPEED 8 tok/s

ⓘ estimated

MEMORY 28/36GB

Gemma 3 27B Q4_K_M

17.0GB 25k cap

Runs comfortably

12 tok/s

ⓘ estimated

17 / 36GB

68 / 71 / 72

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 17/36GB

Qwen3 8B Q8_0

8.5GB 55k cap

Runs comfortably

23 tok/s

ⓘ estimated

9 / 36GB

67 / 70 / 65

FIT Runs comfortably

SPEED 23 tok/s

ⓘ estimated

MEMORY 9/36GB

Qwen3 8B Q4_K_M

5.2GB 61k cap

Runs comfortably

34 tok/s

ⓘ estimated

5 / 36GB

67 / 70 / 65

FIT Runs comfortably

SPEED 34 tok/s

ⓘ estimated

MEMORY 5/36GB

Gemma 3 12B Q8_0

13.5GB 53k cap

Runs comfortably

16 tok/s

ⓘ estimated

14 / 36GB

62 / 65 / 68

FIT Runs comfortably

SPEED 16 tok/s

ⓘ estimated

MEMORY 14/36GB

Mixtral 8x7B Instruct Q3_K_M

19.0GB 32k ctx

Runs comfortably

34 tok/s

ⓘ estimated

19 / 36GB

62 / 65 / 68

FIT Runs comfortably

SPEED 34 tok/s

ⓘ estimated

MEMORY 19/36GB

Mixtral 8x7B Instruct Q4_K_M

26.5GB 19k cap

Runs comfortably

26 tok/s

ⓘ estimated

27 / 36GB

62 / 65 / 68

FIT Runs comfortably

SPEED 26 tok/s

ⓘ estimated

MEMORY 27/36GB

Gemma 3 12B Q4_K_M

7.5GB 67k cap

Runs comfortably

27 tok/s

ⓘ estimated

8 / 36GB

62 / 65 / 68

FIT Runs comfortably

SPEED 27 tok/s

ⓘ estimated

MEMORY 8/36GB

Mistral Nemo 12B Q4_K_M

7.3GB 71k cap

Runs comfortably

28 tok/s

ⓘ estimated

7 / 36GB

58 / 60 / 62

FIT Runs comfortably

SPEED 28 tok/s

ⓘ estimated

MEMORY 7/36GB

Mistral Nemo 12B Q8_0

13.2GB 56k cap

Runs comfortably

17 tok/s

ⓘ estimated

13 / 36GB

58 / 60 / 62

FIT Runs comfortably

SPEED 17 tok/s

ⓘ estimated

MEMORY 13/36GB

Qwen3 4B Q4_K_M

2.6GB 32k ctx

Runs comfortably

66 tok/s

ⓘ estimated

3 / 36GB

58 / 60 / 56

FIT Runs comfortably

SPEED 66 tok/s

ⓘ estimated

MEMORY 3/36GB

Phi-4 Mini Q4_K_M

2.5GB 128k ctx

Runs comfortably

74 tok/s

ⓘ estimated

3 / 36GB

55 / 58 / 52

FIT Runs comfortably

SPEED 74 tok/s

ⓘ estimated

MEMORY 3/36GB

Phi-4 Mini Q8_0

4.1GB 128k ctx

Runs comfortably

50 tok/s

ⓘ estimated

4 / 36GB

55 / 58 / 52

FIT Runs comfortably

SPEED 50 tok/s

ⓘ estimated

MEMORY 4/36GB

Llama 3.1 8B Instruct Q4_K_M

4.5GB 63k cap

Runs comfortably

38 tok/s

ⓘ estimated

5 / 36GB

52 / 55 / 62

FIT Runs comfortably

SPEED 38 tok/s

ⓘ estimated

MEMORY 5/36GB

Llama 3.1 8B Instruct Q8_0

8.0GB 56k cap

Runs comfortably

24 tok/s

ⓘ estimated

8 / 36GB

52 / 55 / 62

FIT Runs comfortably

SPEED 24 tok/s

ⓘ estimated

MEMORY 8/36GB

Llama 3.1 8B Instruct FP16

16.0GB 40k cap

Runs comfortably

14 tok/s

ⓘ estimated

16 / 36GB

52 / 55 / 62

FIT Runs comfortably

SPEED 14 tok/s

ⓘ estimated

MEMORY 16/36GB

Gemma 3 4B Q4_K_M

3.0GB 110k cap

Runs comfortably

58 tok/s

ⓘ estimated

3 / 36GB

42 / 45 / 55

FIT Runs comfortably

SPEED 58 tok/s

ⓘ estimated

MEMORY 3/36GB

Gemma 3 4B Q8_0

5.0GB 103k cap

Runs comfortably

40 tok/s

ⓘ estimated

5 / 36GB

42 / 45 / 55

FIT Runs comfortably

SPEED 40 tok/s

ⓘ estimated

MEMORY 5/36GB

Llama 3.2 3B Instruct Q8_0

3.5GB 128k ctx

Runs comfortably

55 tok/s

ⓘ estimated

4 / 36GB

35 / 38 / 50

FIT Runs comfortably

SPEED 55 tok/s

ⓘ estimated

MEMORY 4/36GB

Llama 3.2 3B Instruct Q4_K_M

2.0GB 128k ctx

Runs comfortably

82 tok/s

ⓘ estimated

2 / 36GB

35 / 38 / 50

FIT Runs comfortably

SPEED 82 tok/s

ⓘ estimated

MEMORY 2/36GB

Mellum2.1 12B Thinking Q4_K_M

7.0GB 131k ctx

Runs comfortably

148 tok/s

ⓘ estimated

7 / 36GB

FIT Runs comfortably

SPEED 148 tok/s

ⓘ estimated

MEMORY 7/36GB

Mellum2.1 12B Thinking FP16

24.3GB 131k ctx

Runs comfortably

47 tok/s

ⓘ estimated

24 / 36GB

FIT Runs comfortably

SPEED 47 tok/s

ⓘ estimated

MEMORY 24/36GB

Muse Glimmer 30B Q4_K_M

18.0GB 20k cap

Runs comfortably

12 tok/s

ⓘ estimated

18 / 36GB

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 18/36GB

LFM2.5-24B-A2B FP8

24.0GB 128k cap

Runs comfortably

106 tok/s

ⓘ estimated

24 / 36GB

FIT Runs comfortably

SPEED 106 tok/s

ⓘ estimated

MEMORY 24/36GB

LFM2.5-24B-A2B Q4_K_M

13.7GB 131k ctx

Runs comfortably

168 tok/s

ⓘ estimated

14 / 36GB

FIT Runs comfortably

SPEED 168 tok/s

ⓘ estimated

MEMORY 14/36GB

Runs tight

Gemma 4 31B Q8_0

32.6GB 2k cap

Runs tight

7 tok/s

ⓘ estimated

33 / 36GB

86 / 89 / 85

FIT Runs tight

SPEED 7 tok/s

ⓘ estimated

MEMORY 33/36GB

Qwen3 32B Q8_0

34.0GB 2k cap

Runs tight

7 tok/s

ⓘ estimated

34 / 36GB

80 / 82 / 78

FIT Runs tight

SPEED 7 tok/s

ⓘ estimated

MEMORY 34/36GB

Nemotron 3.5 Lightning Q4_K_M

32.0GB 4k cap

Runs tight

45 tok/s

ⓘ estimated

32 / 36GB

78 / 76 / 74

FIT Runs tight

SPEED 45 tok/s

ⓘ estimated

MEMORY 32/36GB

Muse Glimmer 30B Q8_0

33.0GB ctx headroom ~0

Runs tight

7 tok/s

ⓘ estimated

33 / 36GB

FIT Runs tight

SPEED 7 tok/s

ⓘ estimated

MEMORY 33/36GB

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