Tokenstead - Find the right AI model for your hardware

RX 7900 XTX 24GB

68 models / sorted by best_quality

Quality Fastest Least mem Newest

Model Fit Speed Memory Scores / Run

Runs comfortably

Ornith-1.5-35B-A3B Q4_K_M

21.7GB 3k cap

Runs comfortably

130 tok/s

ⓘ estimated

22 / 24GB

91 / 89 / 86

FIT Runs comfortably

SPEED 130 tok/s

ⓘ estimated

MEMORY 22/24GB

Qwen3.6 35B A3B Q4_K_M

20.0GB 7k cap

Runs comfortably

133 tok/s

ⓘ estimated

20 / 24GB

90 / 91 / 87

FIT Runs comfortably

SPEED 133 tok/s

ⓘ estimated

MEMORY 20/24GB

Pokee-Isaac 28B Q4_K_M

16.0GB 10k cap

Runs comfortably

29 tok/s

ⓘ estimated

16 / 24GB

90 / 88 / 88

FIT Runs comfortably

SPEED 29 tok/s

ⓘ estimated

MEMORY 16/24GB

Qwen3.8-27B UD-IQ1_S Unsloth

6.2GB 29k cap

Runs comfortably

63 tok/s

ⓘ estimated

6 / 24GB

88 / 89 / 85

FIT Runs comfortably

SPEED 63 tok/s

ⓘ estimated

MEMORY 6/24GB

Qwen3.8-27B UD-Q2_K_XL Unsloth

9.8GB 23k cap

Runs comfortably

44 tok/s

ⓘ estimated

10 / 24GB

88 / 89 / 85

FIT Runs comfortably

SPEED 44 tok/s

ⓘ estimated

MEMORY 10/24GB

Qwen3.8-27B UD-Q4_K_XL Unsloth

17.6GB 9k cap

Runs comfortably

27 tok/s

ⓘ estimated

18 / 24GB

88 / 89 / 85

FIT Runs comfortably

SPEED 27 tok/s

ⓘ estimated

MEMORY 18/24GB

Gemma 4 31B QAT

18.0GB 6k cap

Runs comfortably

25 tok/s

ⓘ estimated

18 / 24GB

86 / 89 / 85

FIT Runs comfortably

SPEED 25 tok/s

ⓘ estimated

MEMORY 18/24GB

Gemma 4 31B Q4_K_M

18.3GB 5k cap

Runs comfortably

24 tok/s

ⓘ estimated

18 / 24GB

86 / 89 / 85

FIT Runs comfortably

SPEED 24 tok/s

ⓘ estimated

MEMORY 18/24GB

Gemma 4 26B A4B Q4_K_M

13.0GB 11k cap

Runs comfortably

109 tok/s

ⓘ estimated

13 / 24GB

85 / 88 / 84

FIT Runs comfortably

SPEED 109 tok/s

ⓘ estimated

MEMORY 13/24GB

Qwen3.6 27B Q4_K_M

16.5GB 10k cap

Runs comfortably

27 tok/s

ⓘ estimated

17 / 24GB

85 / 87 / 82

FIT Runs comfortably

SPEED 27 tok/s

ⓘ estimated

MEMORY 17/24GB

Qwen3 32B Q4_K_M

20.0GB 5k cap

Runs comfortably

23 tok/s

ⓘ estimated

20 / 24GB

80 / 82 / 78

FIT Runs comfortably

SPEED 23 tok/s

ⓘ estimated

MEMORY 20/24GB

Ornith-1.5-9B Q5_K_M

6.6GB 34k cap

Runs comfortably

61 tok/s

ⓘ estimated

7 / 24GB

78 / 76 / 74

FIT Runs comfortably

SPEED 61 tok/s

ⓘ estimated

MEMORY 7/24GB

Ornith-1.5-9B Q4_K_M

5.8GB 36k cap

Runs comfortably

67 tok/s

ⓘ estimated

6 / 24GB

78 / 76 / 74

FIT Runs comfortably

SPEED 67 tok/s

ⓘ estimated

MEMORY 6/24GB

Ornith-1.5-9B BF16

18.4GB 10k cap

Runs comfortably

26 tok/s

ⓘ estimated

18 / 24GB

78 / 76 / 74

FIT Runs comfortably

SPEED 26 tok/s

ⓘ estimated

MEMORY 18/24GB

Ornith-1.5-9B Q8_0

9.8GB 28k cap

Runs comfortably

45 tok/s

ⓘ estimated

10 / 24GB

78 / 76 / 74

FIT Runs comfortably

SPEED 45 tok/s

ⓘ estimated

MEMORY 10/24GB

Ornith-1.5-9B Q6_K

7.6GB 32k cap

Runs comfortably

55 tok/s

ⓘ estimated

8 / 24GB

78 / 76 / 74

FIT Runs comfortably

SPEED 55 tok/s

ⓘ estimated

MEMORY 8/24GB

Qwen3 30B A3B Q4_K_M

15.5GB 15k cap

Runs comfortably

140 tok/s

ⓘ estimated

16 / 24GB

75 / 78 / 72

FIT Runs comfortably

SPEED 140 tok/s

ⓘ estimated

MEMORY 16/24GB

Phi-4 14B Q4_K_M

8.5GB 16k ctx

Runs comfortably

52 tok/s

ⓘ estimated

9 / 24GB

75 / 78 / 70

FIT Runs comfortably

SPEED 52 tok/s

ⓘ estimated

MEMORY 9/24GB

Phi-4 14B Q8_0

15.0GB 16k ctx

Runs comfortably

32 tok/s

ⓘ estimated

15 / 24GB

75 / 78 / 70

FIT Runs comfortably

SPEED 32 tok/s

ⓘ estimated

MEMORY 15/24GB

Xing4.0-29B-A4B Q4_K_M

20.1GB 7k cap

Runs comfortably

69 tok/s

ⓘ estimated

20 / 24GB

75 / 73 / 72

FIT Runs comfortably

SPEED 69 tok/s

ⓘ estimated

MEMORY 20/24GB

Qwen3 14B Q4_K_M

9.0GB 30k cap

Runs comfortably

48 tok/s

ⓘ estimated

9 / 24GB

72 / 75 / 70

FIT Runs comfortably

SPEED 48 tok/s

ⓘ estimated

MEMORY 9/24GB

Qwen3 14B Q8_0

15.0GB 18k cap

Runs comfortably

31 tok/s

ⓘ estimated

15 / 24GB

72 / 75 / 70

FIT Runs comfortably

SPEED 31 tok/s

ⓘ estimated

MEMORY 15/24GB

Gemma 4 12B Q4_K_M

7.2GB 37k cap

Runs comfortably

58 tok/s

ⓘ estimated

7 / 24GB

70 / 74 / 72

FIT Runs comfortably

SPEED 58 tok/s

ⓘ estimated

MEMORY 7/24GB

Gemma 4 12B Q8_0

13.0GB 24k cap

Runs comfortably

36 tok/s

ⓘ estimated

13 / 24GB

70 / 74 / 72

FIT Runs comfortably

SPEED 36 tok/s

ⓘ estimated

MEMORY 13/24GB

Gemma 3 27B Q4_K_M

17.0GB 9k cap

Runs comfortably

26 tok/s

ⓘ estimated

17 / 24GB

68 / 71 / 72

FIT Runs comfortably

SPEED 26 tok/s

ⓘ estimated

MEMORY 17/24GB

Qwen3 8B Q8_0

8.5GB 31k cap

Runs comfortably

50 tok/s

ⓘ estimated

9 / 24GB

67 / 70 / 65

FIT Runs comfortably

SPEED 50 tok/s

ⓘ estimated

MEMORY 9/24GB

Qwen3 8B Q4_K_M

5.2GB 37k cap

Runs comfortably

73 tok/s

ⓘ estimated

5 / 24GB

67 / 70 / 65

FIT Runs comfortably

SPEED 73 tok/s

ⓘ estimated

MEMORY 5/24GB

Gemma 3 12B Q8_0

13.5GB 25k cap

Runs comfortably

35 tok/s

ⓘ estimated

14 / 24GB

62 / 65 / 68

FIT Runs comfortably

SPEED 35 tok/s

ⓘ estimated

MEMORY 14/24GB

Gemma 3 12B Q4_K_M

7.5GB 39k cap

Runs comfortably

57 tok/s

ⓘ estimated

8 / 24GB

62 / 65 / 68

FIT Runs comfortably

SPEED 57 tok/s

ⓘ estimated

MEMORY 8/24GB

Mixtral 8x7B Instruct Q3_K_M

19.0GB 10k cap

Runs comfortably

72 tok/s

ⓘ estimated

19 / 24GB

62 / 65 / 68

FIT Runs comfortably

SPEED 72 tok/s

ⓘ estimated

MEMORY 19/24GB

Mistral Nemo 12B Q4_K_M

7.3GB 41k cap

Runs comfortably

59 tok/s

ⓘ estimated

7 / 24GB

58 / 60 / 62

FIT Runs comfortably

SPEED 59 tok/s

ⓘ estimated

MEMORY 7/24GB

Mistral Nemo 12B Q8_0

13.2GB 26k cap

Runs comfortably

36 tok/s

ⓘ estimated

13 / 24GB

58 / 60 / 62

FIT Runs comfortably

SPEED 36 tok/s

ⓘ estimated

MEMORY 13/24GB

Qwen3 4B Q4_K_M

2.6GB 32k ctx

Runs comfortably

141 tok/s

ⓘ estimated

3 / 24GB

58 / 60 / 56

FIT Runs comfortably

SPEED 141 tok/s

ⓘ estimated

MEMORY 3/24GB

Phi-4 Mini Q4_K_M

2.5GB 107k cap

Runs comfortably

159 tok/s

ⓘ estimated

3 / 24GB

55 / 58 / 52

FIT Runs comfortably

SPEED 159 tok/s

ⓘ estimated

MEMORY 3/24GB

Phi-4 Mini Q8_0

4.1GB 99k cap

Runs comfortably

107 tok/s

ⓘ estimated

4 / 24GB

55 / 58 / 52

FIT Runs comfortably

SPEED 107 tok/s

ⓘ estimated

MEMORY 4/24GB

Llama 3.1 8B Instruct Q8_0

8.0GB 32k cap

Runs comfortably

53 tok/s

ⓘ estimated

8 / 24GB

52 / 55 / 62

FIT Runs comfortably

SPEED 53 tok/s

ⓘ estimated

MEMORY 8/24GB

Llama 3.1 8B Instruct FP16

16.0GB 16k cap

Runs comfortably

29 tok/s

ⓘ estimated

16 / 24GB

52 / 55 / 62

FIT Runs comfortably

SPEED 29 tok/s

ⓘ estimated

MEMORY 16/24GB

Llama 3.1 8B Instruct Q4_K_M

4.5GB 39k cap

Runs comfortably

81 tok/s

ⓘ estimated

5 / 24GB

52 / 55 / 62

FIT Runs comfortably

SPEED 81 tok/s

ⓘ estimated

MEMORY 5/24GB

Gemma 3 4B Q8_0

5.0GB 63k cap

Runs comfortably

85 tok/s

ⓘ estimated

5 / 24GB

42 / 45 / 55

FIT Runs comfortably

SPEED 85 tok/s

ⓘ estimated

MEMORY 5/24GB

Gemma 3 4B Q4_K_M

3.0GB 70k cap

Runs comfortably

125 tok/s

ⓘ estimated

3 / 24GB

42 / 45 / 55

FIT Runs comfortably

SPEED 125 tok/s

ⓘ estimated

MEMORY 3/24GB

Llama 3.2 3B Instruct Q8_0

3.5GB 85k cap

Runs comfortably

118 tok/s

ⓘ estimated

4 / 24GB

35 / 38 / 50

FIT Runs comfortably

SPEED 118 tok/s

ⓘ estimated

MEMORY 4/24GB

Llama 3.2 3B Instruct Q4_K_M

2.0GB 91k cap

Runs comfortably

177 tok/s

ⓘ estimated

2 / 24GB

35 / 38 / 50

FIT Runs comfortably

SPEED 177 tok/s

ⓘ estimated

MEMORY 2/24GB

Mellum2.1 12B Thinking Q4_K_M

7.0GB 131k ctx

Runs comfortably

317 tok/s

ⓘ estimated

7 / 24GB

FIT Runs comfortably

SPEED 317 tok/s

ⓘ estimated

MEMORY 7/24GB

LFM2.5-24B-A2B Q4_K_M

13.7GB 102k cap

Runs comfortably

361 tok/s

ⓘ estimated

14 / 24GB

FIT Runs comfortably

SPEED 361 tok/s

ⓘ estimated

MEMORY 14/24GB

Runs tight

Llama 3.3 70B Instruct Q2_K

21.5GB 1k cap

Runs tight

19 tok/s

ⓘ estimated

22 / 24GB

72 / 77 / 80

FIT Runs tight

SPEED 19 tok/s

ⓘ estimated

MEMORY 22/24GB

Muse Glimmer 30B Q4_K_M

18.0GB ctx headroom ~0

Runs tight

26 tok/s

ⓘ estimated

18 / 24GB

FIT Runs tight

SPEED 26 tok/s

ⓘ estimated

MEMORY 18/24GB

Runs with CPU offload

Qwen3.8-Flash-Next EXL3_2BPW

62.7GB ctx spills to ram

Runs with CPU offload

0.5 tok/s

ⓘ estimated

63 / 24GB

92 / 92 / 89

FIT Runs with CPU offload

SPEED 0.5 tok/s

ⓘ estimated

MEMORY 63/24GB

Qwen3.8-Flash-Next EXL3_3BPW

85.0GB ctx spills to ram

Runs with CPU offload

0.5 tok/s

ⓘ estimated

85 / 24GB

92 / 92 / 89

FIT Runs with CPU offload

SPEED 0.5 tok/s

ⓘ estimated

MEMORY 85/24GB

Ornith-1.5-35B-A3B Q8_0

37.8GB ctx spills to ram

Runs with CPU offload

0.74 tok/s

ⓘ estimated

38 / 24GB

91 / 89 / 86

FIT Runs with CPU offload

SPEED 0.74 tok/s

ⓘ estimated

MEMORY 38/24GB

Ornith-1.5-35B-A3B Q6_K

29.2GB ctx spills to ram

Runs with CPU offload

1 tok/s

ⓘ estimated

29 / 24GB

91 / 89 / 86

FIT Runs with CPU offload

SPEED 1 tok/s

ⓘ estimated

MEMORY 29/24GB

Ornith-1.5-35B-A3B Q5_K_M

25.3GB ctx spills to ram

Runs with CPU offload

3 tok/s

ⓘ estimated

25 / 24GB

91 / 89 / 86

FIT Runs with CPU offload

SPEED 3 tok/s

ⓘ estimated

MEMORY 25/24GB

Qwen3.6 35B A3B Q8_0

37.0GB ctx spills to ram

Runs with CPU offload

0.8 tok/s

ⓘ estimated

37 / 24GB

90 / 91 / 87

FIT Runs with CPU offload

SPEED 0.8 tok/s

ⓘ estimated

MEMORY 37/24GB

Pokee-Isaac 28B Q8_0

28.0GB ctx spills to ram

Runs with CPU offload

7 tok/s

ⓘ estimated

28 / 24GB

90 / 88 / 88

FIT Runs with CPU offload

SPEED 7 tok/s

ⓘ estimated

MEMORY 28/24GB

Gemma 4 31B Q8_0

32.6GB ctx spills to ram

Runs with CPU offload

5 tok/s

ⓘ estimated

33 / 24GB

86 / 89 / 85

FIT Runs with CPU offload

SPEED 5 tok/s

ⓘ estimated

MEMORY 33/24GB

Gemma 4 26B A4B Q8_0

25.0GB ctx spills to ram

Runs with CPU offload

4 tok/s

ⓘ estimated

25 / 24GB

85 / 88 / 84

FIT Runs with CPU offload

SPEED 4 tok/s

ⓘ estimated

MEMORY 25/24GB

Qwen3.6 27B Q8_0

28.0GB ctx spills to ram

Runs with CPU offload

8 tok/s

ⓘ estimated

28 / 24GB

85 / 87 / 82

FIT Runs with CPU offload

SPEED 8 tok/s

ⓘ estimated

MEMORY 28/24GB

Qwen3 32B Q5_K_M

24.5GB ctx spills to ram

Runs with CPU offload

15 tok/s

ⓘ estimated

25 / 24GB

80 / 82 / 78

FIT Runs with CPU offload

SPEED 15 tok/s

ⓘ estimated

MEMORY 25/24GB

Qwen3 32B Q6_K

28.5GB ctx spills to ram

Runs with CPU offload

8 tok/s

ⓘ estimated

29 / 24GB

80 / 82 / 78

FIT Runs with CPU offload

SPEED 8 tok/s

ⓘ estimated

MEMORY 29/24GB

Qwen3 32B Q8_0

34.0GB ctx spills to ram

Runs with CPU offload

5 tok/s

ⓘ estimated

34 / 24GB

80 / 82 / 78

FIT Runs with CPU offload

SPEED 5 tok/s

ⓘ estimated

MEMORY 34/24GB

Nemotron 3.5 Lightning Q4_K_M

32.0GB ctx spills to ram

Runs with CPU offload

1 tok/s

ⓘ estimated

32 / 24GB

78 / 76 / 74

FIT Runs with CPU offload

SPEED 1 tok/s

ⓘ estimated

MEMORY 32/24GB

Qwen3 30B A3B Q8_0

31.0GB ctx spills to ram

Runs with CPU offload

1 tok/s

ⓘ estimated

31 / 24GB

75 / 78 / 72

FIT Runs with CPU offload

SPEED 1 tok/s

ⓘ estimated

MEMORY 31/24GB

Phi-4 14B FP16

29.0GB ctx spills to ram

Runs with CPU offload

7 tok/s

ⓘ estimated

29 / 24GB

75 / 78 / 70

FIT Runs with CPU offload

SPEED 7 tok/s

ⓘ estimated

MEMORY 29/24GB

Llama 3.3 70B Instruct Q3_K_M

30.0GB ctx spills to ram

Runs with CPU offload

7 tok/s

ⓘ estimated

30 / 24GB

72 / 77 / 80

FIT Runs with CPU offload

SPEED 7 tok/s

ⓘ estimated

MEMORY 30/24GB

Gemma 3 27B Q8_0

28.0GB ctx spills to ram

Runs with CPU offload

8 tok/s

ⓘ estimated

28 / 24GB

68 / 71 / 72

FIT Runs with CPU offload

SPEED 8 tok/s

ⓘ estimated

MEMORY 28/24GB

Mixtral 8x7B Instruct Q4_K_M

26.5GB ctx spills to ram

Runs with CPU offload

6 tok/s

ⓘ estimated

27 / 24GB

62 / 65 / 68

FIT Runs with CPU offload

SPEED 6 tok/s

ⓘ estimated

MEMORY 27/24GB

Mellum2.1 12B Thinking FP16

24.3GB ctx spills to ram

Runs with CPU offload

4 tok/s

ⓘ estimated

24 / 24GB

FIT Runs with CPU offload

SPEED 4 tok/s

ⓘ estimated

MEMORY 24/24GB

Muse Glimmer 30B Q8_0

33.0GB ctx spills to ram

Runs with CPU offload

5 tok/s

ⓘ estimated

33 / 24GB

FIT Runs with CPU offload

SPEED 5 tok/s

ⓘ estimated

MEMORY 33/24GB

LFM2.5-24B-A2B FP8

24.0GB ctx spills to ram

Runs with CPU offload

2 tok/s

ⓘ estimated

24 / 24GB

FIT Runs with CPU offload

SPEED 2 tok/s

ⓘ estimated

MEMORY 24/24GB

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