Tokenstead - Find the right AI model for your hardware

M5 Max 48GB

70 models / sorted by best_quality

Quality Fastest Least mem Newest

Model Fit Speed Memory Scores / Run

Runs comfortably

Ornith-1.5-35B-A3B Q8_0

37.8GB 18k cap

Runs comfortably

68 tok/s

ⓘ estimated

38 / 48GB

91 / 89 / 86

FIT Runs comfortably

SPEED 68 tok/s

ⓘ estimated

MEMORY 38/48GB

Ornith-1.5-35B-A3B Q4_K_M

21.7GB 47k cap

Runs comfortably

91 tok/s

ⓘ estimated

22 / 48GB

91 / 89 / 86

FIT Runs comfortably

SPEED 91 tok/s

ⓘ estimated

MEMORY 22/48GB

Ornith-1.5-35B-A3B Q6_K

29.2GB 32k cap

Runs comfortably

79 tok/s

ⓘ estimated

29 / 48GB

91 / 89 / 86

FIT Runs comfortably

SPEED 79 tok/s

ⓘ estimated

MEMORY 29/48GB

Ornith-1.5-35B-A3B Q5_K_M

25.3GB 40k cap

Runs comfortably

84 tok/s

ⓘ estimated

25 / 48GB

91 / 89 / 86

FIT Runs comfortably

SPEED 84 tok/s

ⓘ estimated

MEMORY 25/48GB

Qwen3.6 35B A3B Q8_0

37.0GB 20k cap

Runs comfortably

68 tok/s

ⓘ estimated

37 / 48GB

90 / 91 / 87

FIT Runs comfortably

SPEED 68 tok/s

ⓘ estimated

MEMORY 37/48GB

Qwen3.6 35B A3B Q4_K_M

20.0GB 50k cap

Runs comfortably

93 tok/s

ⓘ estimated

20 / 48GB

90 / 91 / 87

FIT Runs comfortably

SPEED 93 tok/s

ⓘ estimated

MEMORY 20/48GB

Pokee-Isaac 28B Q4_K_M

16.0GB 50k cap

Runs comfortably

20 tok/s

ⓘ estimated

16 / 48GB

90 / 88 / 88

FIT Runs comfortably

SPEED 20 tok/s

ⓘ estimated

MEMORY 16/48GB

Pokee-Isaac 28B Q8_0

28.0GB 30k cap

Runs comfortably

12 tok/s

ⓘ estimated

28 / 48GB

90 / 88 / 88

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 28/48GB

Qwen3.8-27B UD-Q4_K_XL Unsloth

17.6GB 52k cap

Runs comfortably

19 tok/s

ⓘ estimated

18 / 48GB

88 / 89 / 85

FIT Runs comfortably

SPEED 19 tok/s

ⓘ estimated

MEMORY 18/48GB

Qwen3.8-27B UD-Q2_K_XL Unsloth

9.8GB 67k cap

Runs comfortably

30 tok/s

ⓘ estimated

10 / 48GB

88 / 89 / 85

FIT Runs comfortably

SPEED 30 tok/s

ⓘ estimated

MEMORY 10/48GB

Qwen3.8-27B UD-IQ1_S Unsloth

6.2GB 72k cap

Runs comfortably

44 tok/s

ⓘ estimated

6 / 48GB

88 / 89 / 85

FIT Runs comfortably

SPEED 44 tok/s

ⓘ estimated

MEMORY 6/48GB

Qwen3.8-27B NVFP4 Unsloth

15.2GB 54k cap

Runs comfortably

21 tok/s

ⓘ estimated

15 / 48GB

88 / 89 / 85

FIT Runs comfortably

SPEED 21 tok/s

ⓘ estimated

MEMORY 15/48GB

Gemma 4 31B QAT

18.0GB 36k cap

Runs comfortably

17 tok/s

ⓘ estimated

18 / 48GB

86 / 89 / 85

FIT Runs comfortably

SPEED 17 tok/s

ⓘ estimated

MEMORY 18/48GB

Gemma 4 31B NVFP4

16.0GB 35k cap

Runs comfortably

19 tok/s

ⓘ estimated

16 / 48GB

86 / 89 / 85

FIT Runs comfortably

SPEED 19 tok/s

ⓘ estimated

MEMORY 16/48GB

Gemma 4 31B Q8_0

32.6GB 17k cap

Runs comfortably

10 tok/s

ⓘ estimated

33 / 48GB

86 / 89 / 85

FIT Runs comfortably

SPEED 10 tok/s

ⓘ estimated

MEMORY 33/48GB

Gemma 4 31B Q4_K_M

18.3GB 35k cap

Runs comfortably

17 tok/s

ⓘ estimated

18 / 48GB

86 / 89 / 85

FIT Runs comfortably

SPEED 17 tok/s

ⓘ estimated

MEMORY 18/48GB

Gemma 4 26B A4B Q8_0

25.0GB 29k cap

Runs comfortably

56 tok/s

ⓘ estimated

25 / 48GB

85 / 88 / 84

FIT Runs comfortably

SPEED 56 tok/s

ⓘ estimated

MEMORY 25/48GB

Gemma 4 26B A4B Q4_K_M

13.0GB 45k cap

Runs comfortably

76 tok/s

ⓘ estimated

13 / 48GB

85 / 88 / 84

FIT Runs comfortably

SPEED 76 tok/s

ⓘ estimated

MEMORY 13/48GB

Gemma 4 26B A4B NVFP4

14.0GB 47k cap

Runs comfortably

74 tok/s

ⓘ estimated

14 / 48GB

85 / 88 / 84

FIT Runs comfortably

SPEED 74 tok/s

ⓘ estimated

MEMORY 14/48GB

Qwen3.6 27B Q4_K_M

16.5GB 44k cap

Runs comfortably

19 tok/s

ⓘ estimated

17 / 48GB

85 / 87 / 82

FIT Runs comfortably

SPEED 19 tok/s

ⓘ estimated

MEMORY 17/48GB

Qwen3.6 27B Q8_0

28.0GB 28k cap

Runs comfortably

12 tok/s

ⓘ estimated

28 / 48GB

85 / 87 / 82

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 28/48GB

Qwen3 32B Q4_K_M

20.0GB 39k cap

Runs comfortably

16 tok/s

ⓘ estimated

20 / 48GB

80 / 82 / 78

FIT Runs comfortably

SPEED 16 tok/s

ⓘ estimated

MEMORY 20/48GB

Qwen3 32B Q5_K_M

24.5GB 33k cap

Runs comfortably

13 tok/s

ⓘ estimated

25 / 48GB

80 / 82 / 78

FIT Runs comfortably

SPEED 13 tok/s

ⓘ estimated

MEMORY 25/48GB

Qwen3 32B Q6_K

28.5GB 27k cap

Runs comfortably

12 tok/s

ⓘ estimated

29 / 48GB

80 / 82 / 78

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 29/48GB

Qwen3 32B Q8_0

34.0GB 19k cap

Runs comfortably

10 tok/s

ⓘ estimated

34 / 48GB

80 / 82 / 78

FIT Runs comfortably

SPEED 10 tok/s

ⓘ estimated

MEMORY 34/48GB

Ornith-1.5-9B Q6_K

7.6GB 80k cap

Runs comfortably

38 tok/s

ⓘ estimated

8 / 48GB

78 / 76 / 74

FIT Runs comfortably

SPEED 38 tok/s

ⓘ estimated

MEMORY 8/48GB

Nemotron 3.5 Lightning Q4_K_M

32.0GB 31k cap

Runs comfortably

67 tok/s

ⓘ estimated

32 / 48GB

78 / 76 / 74

FIT Runs comfortably

SPEED 67 tok/s

ⓘ estimated

MEMORY 32/48GB

Ornith-1.5-9B Q8_0

9.8GB 76k cap

Runs comfortably

31 tok/s

ⓘ estimated

10 / 48GB

78 / 76 / 74

FIT Runs comfortably

SPEED 31 tok/s

ⓘ estimated

MEMORY 10/48GB

Ornith-1.5-9B BF16

18.4GB 58k cap

Runs comfortably

18 tok/s

ⓘ estimated

18 / 48GB

78 / 76 / 74

FIT Runs comfortably

SPEED 18 tok/s

ⓘ estimated

MEMORY 18/48GB

Ornith-1.5-9B Q4_K_M

5.8GB 84k cap

Runs comfortably

47 tok/s

ⓘ estimated

6 / 48GB

78 / 76 / 74

FIT Runs comfortably

SPEED 47 tok/s

ⓘ estimated

MEMORY 6/48GB

Ornith-1.5-9B Q5_K_M

6.6GB 82k cap

Runs comfortably

42 tok/s

ⓘ estimated

7 / 48GB

78 / 76 / 74

FIT Runs comfortably

SPEED 42 tok/s

ⓘ estimated

MEMORY 7/48GB

Qwen3 30B A3B Q8_0

31.0GB 30k cap

Runs comfortably

69 tok/s

ⓘ estimated

31 / 48GB

75 / 78 / 72

FIT Runs comfortably

SPEED 69 tok/s

ⓘ estimated

MEMORY 31/48GB

Qwen3 30B A3B Q4_K_M

15.5GB 59k cap

Runs comfortably

98 tok/s

ⓘ estimated

16 / 48GB

75 / 78 / 72

FIT Runs comfortably

SPEED 98 tok/s

ⓘ estimated

MEMORY 16/48GB

Phi-4 14B Q8_0

15.0GB 16k ctx

Runs comfortably

22 tok/s

ⓘ estimated

15 / 48GB

75 / 78 / 70

FIT Runs comfortably

SPEED 22 tok/s

ⓘ estimated

MEMORY 15/48GB

Phi-4 14B FP16

29.0GB 16k ctx

Runs comfortably

12 tok/s

ⓘ estimated

29 / 48GB

75 / 78 / 70

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 29/48GB

Phi-4 14B Q4_K_M

8.5GB 16k ctx

Runs comfortably

36 tok/s

ⓘ estimated

9 / 48GB

75 / 78 / 70

FIT Runs comfortably

SPEED 36 tok/s

ⓘ estimated

MEMORY 9/48GB

Xing4.0-29B-A4B Q4_K_M

20.1GB 27k cap

Runs comfortably

48 tok/s

ⓘ estimated

20 / 48GB

75 / 73 / 72

FIT Runs comfortably

SPEED 48 tok/s

ⓘ estimated

MEMORY 20/48GB

Llama 3.3 70B Instruct Q4_K_M

42.0GB 4k cap

Runs comfortably

8 tok/s

ⓘ estimated

42 / 48GB

72 / 77 / 80

FIT Runs comfortably

SPEED 8 tok/s

ⓘ estimated

MEMORY 42/48GB

Llama 3.3 70B Instruct Q3_K_M

30.0GB 12k cap

Runs comfortably

10 tok/s

ⓘ estimated

30 / 48GB

72 / 77 / 80

FIT Runs comfortably

SPEED 10 tok/s

ⓘ estimated

MEMORY 30/48GB

Llama 3.3 70B Instruct Q2_K

21.5GB 17k cap

Runs comfortably

13 tok/s

ⓘ estimated

22 / 48GB

72 / 77 / 80

FIT Runs comfortably

SPEED 13 tok/s

ⓘ estimated

MEMORY 22/48GB

Qwen3 14B Q4_K_M

9.0GB 78k cap

Runs comfortably

33 tok/s

ⓘ estimated

9 / 48GB

72 / 75 / 70

FIT Runs comfortably

SPEED 33 tok/s

ⓘ estimated

MEMORY 9/48GB

Qwen3 14B Q8_0

15.0GB 66k cap

Runs comfortably

22 tok/s

ⓘ estimated

15 / 48GB

72 / 75 / 70

FIT Runs comfortably

SPEED 22 tok/s

ⓘ estimated

MEMORY 15/48GB

Gemma 4 12B Q4_K_M

7.2GB 90k cap

Runs comfortably

41 tok/s

ⓘ estimated

7 / 48GB

70 / 74 / 72

FIT Runs comfortably

SPEED 41 tok/s

ⓘ estimated

MEMORY 7/48GB

Gemma 4 12B Q8_0

13.0GB 77k cap

Runs comfortably

25 tok/s

ⓘ estimated

13 / 48GB

70 / 74 / 72

FIT Runs comfortably

SPEED 25 tok/s

ⓘ estimated

MEMORY 13/48GB

Gemma 3 27B Q8_0

28.0GB 26k cap

Runs comfortably

12 tok/s

ⓘ estimated

28 / 48GB

68 / 71 / 72

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 28/48GB

Gemma 3 27B Q4_K_M

17.0GB 41k cap

Runs comfortably

18 tok/s

ⓘ estimated

17 / 48GB

68 / 71 / 72

FIT Runs comfortably

SPEED 18 tok/s

ⓘ estimated

MEMORY 17/48GB

Qwen3 8B Q4_K_M

5.2GB 85k cap

Runs comfortably

51 tok/s

ⓘ estimated

5 / 48GB

67 / 70 / 65

FIT Runs comfortably

SPEED 51 tok/s

ⓘ estimated

MEMORY 5/48GB

Qwen3 8B Q8_0

8.5GB 79k cap

Runs comfortably

35 tok/s

ⓘ estimated

9 / 48GB

67 / 70 / 65

FIT Runs comfortably

SPEED 35 tok/s

ⓘ estimated

MEMORY 9/48GB

Gemma 3 12B Q4_K_M

7.5GB 96k cap

Runs comfortably

40 tok/s

ⓘ estimated

8 / 48GB

62 / 65 / 68

FIT Runs comfortably

SPEED 40 tok/s

ⓘ estimated

MEMORY 8/48GB

Gemma 3 12B Q8_0

13.5GB 82k cap

Runs comfortably

24 tok/s

ⓘ estimated

14 / 48GB

62 / 65 / 68

FIT Runs comfortably

SPEED 24 tok/s

ⓘ estimated

MEMORY 14/48GB

Mixtral 8x7B Instruct Q3_K_M

19.0GB 32k ctx

Runs comfortably

50 tok/s

ⓘ estimated

19 / 48GB

62 / 65 / 68

FIT Runs comfortably

SPEED 50 tok/s

ⓘ estimated

MEMORY 19/48GB

Mixtral 8x7B Instruct Q4_K_M

26.5GB 32k ctx

Runs comfortably

39 tok/s

ⓘ estimated

27 / 48GB

62 / 65 / 68

FIT Runs comfortably

SPEED 39 tok/s

ⓘ estimated

MEMORY 27/48GB

Mistral Nemo 12B Q4_K_M

7.3GB 101k cap

Runs comfortably

41 tok/s

ⓘ estimated

7 / 48GB

58 / 60 / 62

FIT Runs comfortably

SPEED 41 tok/s

ⓘ estimated

MEMORY 7/48GB

Mistral Nemo 12B Q8_0

13.2GB 86k cap

Runs comfortably

25 tok/s

ⓘ estimated

13 / 48GB

58 / 60 / 62

FIT Runs comfortably

SPEED 25 tok/s

ⓘ estimated

MEMORY 13/48GB

Qwen3 4B Q4_K_M

2.6GB 32k ctx

Runs comfortably

98 tok/s

ⓘ estimated

3 / 48GB

58 / 60 / 56

FIT Runs comfortably

SPEED 98 tok/s

ⓘ estimated

MEMORY 3/48GB

Phi-4 Mini Q4_K_M

2.5GB 128k ctx

Runs comfortably

111 tok/s

ⓘ estimated

3 / 48GB

55 / 58 / 52

FIT Runs comfortably

SPEED 111 tok/s

ⓘ estimated

MEMORY 3/48GB

Phi-4 Mini Q8_0

4.1GB 128k ctx

Runs comfortably

75 tok/s

ⓘ estimated

4 / 48GB

55 / 58 / 52

FIT Runs comfortably

SPEED 75 tok/s

ⓘ estimated

MEMORY 4/48GB

Llama 3.1 8B Instruct FP16

16.0GB 64k cap

Runs comfortably

20 tok/s

ⓘ estimated

16 / 48GB

52 / 55 / 62

FIT Runs comfortably

SPEED 20 tok/s

ⓘ estimated

MEMORY 16/48GB

Llama 3.1 8B Instruct Q8_0

8.0GB 80k cap

Runs comfortably

37 tok/s

ⓘ estimated

8 / 48GB

52 / 55 / 62

FIT Runs comfortably

SPEED 37 tok/s

ⓘ estimated

MEMORY 8/48GB

Llama 3.1 8B Instruct Q4_K_M

4.5GB 87k cap

Runs comfortably

56 tok/s

ⓘ estimated

5 / 48GB

52 / 55 / 62

FIT Runs comfortably

SPEED 56 tok/s

ⓘ estimated

MEMORY 5/48GB

Gemma 3 4B Q8_0

5.0GB 128k ctx

Runs comfortably

59 tok/s

ⓘ estimated

5 / 48GB

42 / 45 / 55

FIT Runs comfortably

SPEED 59 tok/s

ⓘ estimated

MEMORY 5/48GB

Gemma 3 4B Q4_K_M

3.0GB 128k ctx

Runs comfortably

87 tok/s

ⓘ estimated

3 / 48GB

42 / 45 / 55

FIT Runs comfortably

SPEED 87 tok/s

ⓘ estimated

MEMORY 3/48GB

Llama 3.2 3B Instruct Q8_0

3.5GB 128k ctx

Runs comfortably

82 tok/s

ⓘ estimated

4 / 48GB

35 / 38 / 50

FIT Runs comfortably

SPEED 82 tok/s

ⓘ estimated

MEMORY 4/48GB

Llama 3.2 3B Instruct Q4_K_M

2.0GB 128k ctx

Runs comfortably

123 tok/s

ⓘ estimated

2 / 48GB

35 / 38 / 50

FIT Runs comfortably

SPEED 123 tok/s

ⓘ estimated

MEMORY 2/48GB

Mellum2.1 12B Thinking Q4_K_M

7.0GB 131k ctx

Runs comfortably

221 tok/s

ⓘ estimated

7 / 48GB

FIT Runs comfortably

SPEED 221 tok/s

ⓘ estimated

MEMORY 7/48GB

Mellum2.1 12B Thinking FP16

24.3GB 131k ctx

Runs comfortably

71 tok/s

ⓘ estimated

24 / 48GB

FIT Runs comfortably

SPEED 71 tok/s

ⓘ estimated

MEMORY 24/48GB

Muse Glimmer 30B Q8_0

33.0GB 20k cap

Runs comfortably

10 tok/s

ⓘ estimated

33 / 48GB

FIT Runs comfortably

SPEED 10 tok/s

ⓘ estimated

MEMORY 33/48GB

Muse Glimmer 30B Q4_K_M

18.0GB 40k cap

Runs comfortably

18 tok/s

ⓘ estimated

18 / 48GB

FIT Runs comfortably

SPEED 18 tok/s

ⓘ estimated

MEMORY 18/48GB

LFM2.5-24B-A2B FP8

24.0GB 131k ctx

Runs comfortably

159 tok/s

ⓘ estimated

24 / 48GB

FIT Runs comfortably

SPEED 159 tok/s

ⓘ estimated

MEMORY 24/48GB

LFM2.5-24B-A2B Q4_K_M

13.7GB 131k ctx

Runs comfortably

252 tok/s

ⓘ estimated

14 / 48GB

FIT Runs comfortably

SPEED 252 tok/s

ⓘ estimated

MEMORY 14/48GB

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