Tokenstead - Find the right AI model for your hardware

GB300 (DGX Station) 748GB

110 models / sorted by best_quality

Quality Fastest Least mem Newest

Model Fit Speed Memory Scores / Run

Runs comfortably

GLM-5.3-Flash UD-IQ1_S Unsloth

93.1GB 1000k ctx

Runs comfortably

745 tok/s

ⓘ estimated

93 / 748GB

93 / 92 / 91

FIT Runs comfortably

SPEED 745 tok/s

ⓘ estimated

MEMORY 93/748GB

GLM-5.3-Flash UD-Q2_K_XL Unsloth

109.0GB 1000k ctx

Runs comfortably

636 tok/s

ⓘ estimated

109 / 748GB

93 / 92 / 91

FIT Runs comfortably

SPEED 636 tok/s

ⓘ estimated

MEMORY 109/748GB

GLM-5.3-Flash UD-IQ3_XXS Unsloth

120.0GB 1000k ctx

Runs comfortably

578 tok/s

ⓘ estimated

120 / 748GB

93 / 92 / 91

FIT Runs comfortably

SPEED 578 tok/s

ⓘ estimated

MEMORY 120/748GB

GLM-5.3-Flash UD-Q4_K_XL Unsloth

200.0GB 1000k ctx

Runs comfortably

347 tok/s

ⓘ estimated

200 / 748GB

93 / 92 / 91

FIT Runs comfortably

SPEED 347 tok/s

ⓘ estimated

MEMORY 200/748GB

DeepSeek V4.1 Flash FP8

510.3GB 1000k ctx

Runs comfortably

264 tok/s

ⓘ estimated

510 / 748GB

93 / 92 / 90

FIT Runs comfortably

SPEED 264 tok/s

ⓘ estimated

MEMORY 510/748GB

DeepSeek V4 Flash 0731 UD-Q8_K_XL Unsloth

162.0GB 1000k ctx

Runs comfortably

526 tok/s

ⓘ estimated

162 / 748GB

93 / 91 / 90

FIT Runs comfortably

SPEED 526 tok/s

ⓘ estimated

MEMORY 162/748GB

DeepSeek V4 Flash 0731 UD-Q4_K_XL Unsloth

155.0GB 1000k ctx

Runs comfortably

1300 tok/s

extrapolated

155 / 748GB

93 / 91 / 90

FIT Runs comfortably

SPEED 1300 tok/s

extrapolated

MEMORY 155/748GB

Ornith-1.5-397B Q5_K_M

286.0GB 262k ctx

Runs comfortably

14 tok/s

ⓘ estimated

286 / 748GB

93 / 91 / 89

FIT Runs comfortably

SPEED 14 tok/s

ⓘ estimated

MEMORY 286/748GB

Ornith-1.5-397B Q6_K

331.0GB 262k ctx

Runs comfortably

12 tok/s

ⓘ estimated

331 / 748GB

93 / 91 / 89

FIT Runs comfortably

SPEED 12 tok/s

ⓘ estimated

MEMORY 331/748GB

Ornith-1.5-397B Q8_0

429.0GB 262k ctx

Runs comfortably

9 tok/s

ⓘ estimated

429 / 748GB

93 / 91 / 89

FIT Runs comfortably

SPEED 9 tok/s

ⓘ estimated

MEMORY 429/748GB

Ornith-1.5-397B Q4_K_M

244.0GB 262k ctx

Runs comfortably

16 tok/s

ⓘ estimated

244 / 748GB

93 / 91 / 89

FIT Runs comfortably

SPEED 16 tok/s

ⓘ estimated

MEMORY 244/748GB

Qwen3.8-Flash-Next UD-IQ1_S Unsloth

72.5GB 262k ctx

Runs comfortably

1614 tok/s

ⓘ estimated

73 / 748GB

92 / 92 / 89

FIT Runs comfortably

SPEED 1614 tok/s

ⓘ estimated

MEMORY 73/748GB

Qwen3.8-Flash-Next EXL3_3BPW

85.0GB 262k ctx

Runs comfortably

858 tok/s

ⓘ estimated

85 / 748GB

92 / 92 / 89

FIT Runs comfortably

SPEED 858 tok/s

ⓘ estimated

MEMORY 85/748GB

Qwen3.8-Flash-Next EXL3_2BPW

62.7GB 262k ctx

Runs comfortably

1025 tok/s

ⓘ estimated

63 / 748GB

92 / 92 / 89

FIT Runs comfortably

SPEED 1025 tok/s

ⓘ estimated

MEMORY 63/748GB

GLM 5.2 Q2_K Unsloth

240.0GB 1000k ctx

Runs comfortably

302 tok/s

ⓘ estimated

240 / 748GB

91 / 92 / 88

FIT Runs comfortably

SPEED 302 tok/s

ⓘ estimated

MEMORY 240/748GB

GLM 5.2 Q3_K_M Unsloth

365.0GB 1000k ctx

Runs comfortably

199 tok/s

ⓘ estimated

365 / 748GB

91 / 92 / 88

FIT Runs comfortably

SPEED 199 tok/s

ⓘ estimated

MEMORY 365/748GB

GLM 5.2 Q4_K_M Unsloth

410.0GB 1000k ctx

Runs comfortably

177 tok/s

ⓘ estimated

410 / 748GB

91 / 92 / 88

FIT Runs comfortably

SPEED 177 tok/s

ⓘ estimated

MEMORY 410/748GB

GLM 5.2 Q5_K_M Unsloth

510.0GB 1000k ctx

Runs comfortably

142 tok/s

ⓘ estimated

510 / 748GB

91 / 92 / 88

FIT Runs comfortably

SPEED 142 tok/s

ⓘ estimated

MEMORY 510/748GB

DeepSeek V3 0324 Q2_K

188.0GB 128k ctx

Runs comfortably

376 tok/s

ⓘ estimated

188 / 748GB

92 / 88 / 87

FIT Runs comfortably

SPEED 376 tok/s

ⓘ estimated

MEMORY 188/748GB

DeepSeek V3 0324 Q4_K_M

368.0GB 128k ctx

Runs comfortably

192 tok/s

ⓘ estimated

368 / 748GB

92 / 88 / 87

FIT Runs comfortably

SPEED 192 tok/s

ⓘ estimated

MEMORY 368/748GB

Ornith-1.5-35B-A3B Q8_0

37.8GB 262k ctx

Runs comfortably

1239 tok/s

ⓘ estimated

38 / 748GB

91 / 89 / 86

FIT Runs comfortably

SPEED 1239 tok/s

ⓘ estimated

MEMORY 38/748GB

Ornith-1.5-35B-A3B BF16

71.1GB 262k ctx

Runs comfortably

659 tok/s

ⓘ estimated

71 / 748GB

91 / 89 / 86

FIT Runs comfortably

SPEED 659 tok/s

ⓘ estimated

MEMORY 71/748GB

Ornith-1.5-35B-A3B Q6_K

29.2GB 262k ctx

Runs comfortably

1603 tok/s

ⓘ estimated

29 / 748GB

91 / 89 / 86

FIT Runs comfortably

SPEED 1603 tok/s

ⓘ estimated

MEMORY 29/748GB

Ornith-1.5-35B-A3B Q4_K_M

21.7GB 262k ctx

Runs comfortably

2157 tok/s

ⓘ estimated

22 / 748GB

91 / 89 / 86

FIT Runs comfortably

SPEED 2157 tok/s

ⓘ estimated

MEMORY 22/748GB

Ornith-1.5-35B-A3B Q5_K_M

25.3GB 262k ctx

Runs comfortably

1850 tok/s

ⓘ estimated

25 / 748GB

91 / 89 / 86

FIT Runs comfortably

SPEED 1850 tok/s

ⓘ estimated

MEMORY 25/748GB

Qwen3.6 35B A3B Q8_0

37.0GB 262k ctx

Runs comfortably

1230 tok/s

ⓘ estimated

37 / 748GB

90 / 91 / 87

FIT Runs comfortably

SPEED 1230 tok/s

ⓘ estimated

MEMORY 37/748GB

Qwen3.6 35B A3B Q4_K_M

20.0GB 262k ctx

Runs comfortably

2275 tok/s

ⓘ estimated

20 / 748GB

90 / 91 / 87

FIT Runs comfortably

SPEED 2275 tok/s

ⓘ estimated

MEMORY 20/748GB

Pokee-Isaac 28B Q4_K_M

16.0GB 10000k ctx

Runs comfortably

244 tok/s

ⓘ estimated

16 / 748GB

90 / 88 / 88

FIT Runs comfortably

SPEED 244 tok/s

ⓘ estimated

MEMORY 16/748GB

Pokee-Isaac 28B Q8_0

28.0GB 10000k ctx

Runs comfortably

139 tok/s

ⓘ estimated

28 / 748GB

90 / 88 / 88

FIT Runs comfortably

SPEED 139 tok/s

ⓘ estimated

MEMORY 28/748GB

Pokee-Isaac 28B FP16

56.0GB 10000k ctx

Runs comfortably

70 tok/s

ⓘ estimated

56 / 748GB

90 / 88 / 88

FIT Runs comfortably

SPEED 70 tok/s

ⓘ estimated

MEMORY 56/748GB

Qwen3 235B A22B Q4_K_M

125.0GB 128k ctx

Runs comfortably

333 tok/s

ⓘ estimated

125 / 748GB

88 / 90 / 85

FIT Runs comfortably

SPEED 333 tok/s

ⓘ estimated

MEMORY 125/748GB

Qwen3 235B A22B Q2_K

68.0GB 128k ctx

Runs comfortably

613 tok/s

ⓘ estimated

68 / 748GB

88 / 90 / 85

FIT Runs comfortably

SPEED 613 tok/s

ⓘ estimated

MEMORY 68/748GB

Qwen3.8-27B UD-IQ1_S Unsloth

6.2GB 262k ctx

Runs comfortably

631 tok/s

ⓘ estimated

6 / 748GB

88 / 89 / 85

FIT Runs comfortably

SPEED 631 tok/s

ⓘ estimated

MEMORY 6/748GB

Qwen3.8-27B UD-Q2_K_XL Unsloth

9.8GB 262k ctx

Runs comfortably

397 tok/s

ⓘ estimated

10 / 748GB

88 / 89 / 85

FIT Runs comfortably

SPEED 397 tok/s

ⓘ estimated

MEMORY 10/748GB

Qwen3.8-27B UD-Q4_K_XL Unsloth

17.6GB 262k ctx

Runs comfortably

222 tok/s

ⓘ estimated

18 / 748GB

88 / 89 / 85

FIT Runs comfortably

SPEED 222 tok/s

ⓘ estimated

MEMORY 18/748GB

Gemma 4 31B Q4_K_M

18.3GB 256k ctx

Runs comfortably

213 tok/s

ⓘ estimated

18 / 748GB

86 / 89 / 85

FIT Runs comfortably

SPEED 213 tok/s

ⓘ estimated

MEMORY 18/748GB

Gemma 4 31B Q8_0

32.6GB 256k ctx

Runs comfortably

120 tok/s

ⓘ estimated

33 / 748GB

86 / 89 / 85

FIT Runs comfortably

SPEED 120 tok/s

ⓘ estimated

MEMORY 33/748GB

Gemma 4 31B BF16

61.4GB 256k ctx

Runs comfortably

64 tok/s

ⓘ estimated

61 / 748GB

86 / 89 / 85

FIT Runs comfortably

SPEED 64 tok/s

ⓘ estimated

MEMORY 61/748GB

Gemma 4 31B QAT

18.0GB 256k ctx

Runs comfortably

217 tok/s

ⓘ estimated

18 / 748GB

86 / 89 / 85

FIT Runs comfortably

SPEED 217 tok/s

ⓘ estimated

MEMORY 18/748GB

Gemma 4 26B A4B Q4_K_M

13.0GB 256k ctx

Runs comfortably

1989 tok/s

ⓘ estimated

13 / 748GB

85 / 88 / 84

FIT Runs comfortably

SPEED 1989 tok/s

ⓘ estimated

MEMORY 13/748GB

Gemma 4 26B A4B Q8_0

25.0GB 256k ctx

Runs comfortably

1035 tok/s

ⓘ estimated

25 / 748GB

85 / 88 / 84

FIT Runs comfortably

SPEED 1035 tok/s

ⓘ estimated

MEMORY 25/748GB

Gemma 4 26B A4B BF16

50.0GB 256k ctx

Runs comfortably

518 tok/s

ⓘ estimated

50 / 748GB

85 / 88 / 84

FIT Runs comfortably

SPEED 518 tok/s

ⓘ estimated

MEMORY 50/748GB

Qwen3.6 27B Q4_K_M

16.5GB 128k ctx

Runs comfortably

237 tok/s

ⓘ estimated

17 / 748GB

85 / 87 / 82

FIT Runs comfortably

SPEED 237 tok/s

ⓘ estimated

MEMORY 17/748GB

Qwen3.6 27B Q8_0

28.0GB 128k ctx

Runs comfortably

139 tok/s

ⓘ estimated

28 / 748GB

85 / 87 / 82

FIT Runs comfortably

SPEED 139 tok/s

ⓘ estimated

MEMORY 28/748GB

Qwen3 32B Q4_K_M

20.0GB 128k ctx

Runs comfortably

195 tok/s

ⓘ estimated

20 / 748GB

80 / 82 / 78

FIT Runs comfortably

SPEED 195 tok/s

ⓘ estimated

MEMORY 20/748GB

Qwen3 32B Q8_0

34.0GB 128k ctx

Runs comfortably

115 tok/s

ⓘ estimated

34 / 748GB

80 / 82 / 78

FIT Runs comfortably

SPEED 115 tok/s

ⓘ estimated

MEMORY 34/748GB

Qwen3 32B Q5_K_M

24.5GB 128k ctx

Runs comfortably

159 tok/s

ⓘ estimated

25 / 748GB

80 / 82 / 78

FIT Runs comfortably

SPEED 159 tok/s

ⓘ estimated

MEMORY 25/748GB

Qwen3 32B Q6_K

28.5GB 128k ctx

Runs comfortably

137 tok/s

ⓘ estimated

29 / 748GB

80 / 82 / 78

FIT Runs comfortably

SPEED 137 tok/s

ⓘ estimated

MEMORY 29/748GB

MiMo-V2.6-Flash Q2_K

77.0GB 167k cap

Runs comfortably

194 tok/s

ⓘ estimated

77 / 748GB

78 / 77 / 76

FIT Runs comfortably

SPEED 194 tok/s

ⓘ estimated

MEMORY 77/748GB

MiMo-V2.6-Flash FP8

177.7GB 142k cap

Runs comfortably

156 tok/s

ⓘ estimated

178 / 748GB

78 / 77 / 76

FIT Runs comfortably

SPEED 156 tok/s

ⓘ estimated

MEMORY 178/748GB

Ornith-1.5-9B Q8_0

9.8GB 262k ctx

Runs comfortably

399 tok/s

ⓘ estimated

10 / 748GB

78 / 76 / 74

FIT Runs comfortably

SPEED 399 tok/s

ⓘ estimated

MEMORY 10/748GB

Ornith-1.5-9B BF16

18.4GB 262k ctx

Runs comfortably

212 tok/s

ⓘ estimated

18 / 748GB

78 / 76 / 74

FIT Runs comfortably

SPEED 212 tok/s

ⓘ estimated

MEMORY 18/748GB

Ornith-1.5-9B Q6_K

7.6GB 262k ctx

Runs comfortably

516 tok/s

ⓘ estimated

8 / 748GB

78 / 76 / 74

FIT Runs comfortably

SPEED 516 tok/s

ⓘ estimated

MEMORY 8/748GB

Ornith-1.5-9B Q5_K_M

6.6GB 262k ctx

Runs comfortably

588 tok/s

ⓘ estimated

7 / 748GB

78 / 76 / 74

FIT Runs comfortably

SPEED 588 tok/s

ⓘ estimated

MEMORY 7/748GB

Ornith-1.5-9B Q4_K_M

5.8GB 262k ctx

Runs comfortably

675 tok/s

ⓘ estimated

6 / 748GB

78 / 76 / 74

FIT Runs comfortably

SPEED 675 tok/s

ⓘ estimated

MEMORY 6/748GB

Nemotron 3.5 Lightning Q4_K_M

32.0GB 1000k ctx

Runs comfortably

1071 tok/s

ⓘ estimated

32 / 748GB

78 / 76 / 74

FIT Runs comfortably

SPEED 1071 tok/s

ⓘ estimated

MEMORY 32/748GB

Nemotron 3.5 Lightning Q8_0

58.0GB 1000k ctx

Runs comfortably

591 tok/s

ⓘ estimated

58 / 748GB

78 / 76 / 74

FIT Runs comfortably

SPEED 591 tok/s

ⓘ estimated

MEMORY 58/748GB

Nemotron 3.5 Lightning FP16

60.0GB 1000k ctx

Runs comfortably

571 tok/s

ⓘ estimated

60 / 748GB

78 / 76 / 74

FIT Runs comfortably

SPEED 571 tok/s

ⓘ estimated

MEMORY 60/748GB

Qwen3 30B A3B Q8_0

31.0GB 128k ctx

Runs comfortably

1280 tok/s

ⓘ estimated

31 / 748GB

75 / 78 / 72

FIT Runs comfortably

SPEED 1280 tok/s

ⓘ estimated

MEMORY 31/748GB

Qwen3 30B A3B Q4_K_M

15.5GB 128k ctx

Runs comfortably

2558 tok/s

ⓘ estimated

16 / 748GB

75 / 78 / 72

FIT Runs comfortably

SPEED 2558 tok/s

ⓘ estimated

MEMORY 16/748GB

Phi-4 14B Q4_K_M

8.5GB 16k ctx

Runs comfortably

459 tok/s

ⓘ estimated

9 / 748GB

75 / 78 / 70

FIT Runs comfortably

SPEED 459 tok/s

ⓘ estimated

MEMORY 9/748GB

Phi-4 14B Q8_0

15.0GB 16k ctx

Runs comfortably

260 tok/s

ⓘ estimated

15 / 748GB

75 / 78 / 70

FIT Runs comfortably

SPEED 260 tok/s

ⓘ estimated

MEMORY 15/748GB

Phi-4 14B FP16

29.0GB 16k ctx

Runs comfortably

135 tok/s

ⓘ estimated

29 / 748GB

75 / 78 / 70

FIT Runs comfortably

SPEED 135 tok/s

ⓘ estimated

MEMORY 29/748GB

Xing4.0-29B-A4B BF16

62.4GB 256k ctx

Runs comfortably

289 tok/s

ⓘ estimated

62 / 748GB

75 / 73 / 72

FIT Runs comfortably

SPEED 289 tok/s

ⓘ estimated

MEMORY 62/748GB

Xing4.0-29B-A4B Q4_K_M

20.1GB 256k ctx

Runs comfortably

508 tok/s

ⓘ estimated

20 / 748GB

75 / 73 / 72

FIT Runs comfortably

SPEED 508 tok/s

ⓘ estimated

MEMORY 20/748GB

Llama 3.3 70B Instruct Q4_K_M

42.0GB 128k ctx

Runs comfortably

93 tok/s

ⓘ estimated

42 / 748GB

72 / 77 / 80

FIT Runs comfortably

SPEED 93 tok/s

ⓘ estimated

MEMORY 42/748GB

Llama 3.3 70B Instruct Q8_0

74.0GB 128k ctx

Runs comfortably

53 tok/s

ⓘ estimated

74 / 748GB

72 / 77 / 80

FIT Runs comfortably

SPEED 53 tok/s

ⓘ estimated

MEMORY 74/748GB

Llama 3.3 70B Instruct Q3_K_M

30.0GB 128k ctx

Runs comfortably

130 tok/s

ⓘ estimated

30 / 748GB

72 / 77 / 80

FIT Runs comfortably

SPEED 130 tok/s

ⓘ estimated

MEMORY 30/748GB

Llama 3.3 70B Instruct Q2_K

21.5GB 128k ctx

Runs comfortably

182 tok/s

ⓘ estimated

22 / 748GB

72 / 77 / 80

FIT Runs comfortably

SPEED 182 tok/s

ⓘ estimated

MEMORY 22/748GB

Qwen3 14B Q4_K_M

9.0GB 128k ctx

Runs comfortably

434 tok/s

ⓘ estimated

9 / 748GB

72 / 75 / 70

FIT Runs comfortably

SPEED 434 tok/s

ⓘ estimated

MEMORY 9/748GB

Qwen3 14B Q8_0

15.0GB 128k ctx

Runs comfortably

260 tok/s

ⓘ estimated

15 / 748GB

72 / 75 / 70

FIT Runs comfortably

SPEED 260 tok/s

ⓘ estimated

MEMORY 15/748GB

Gemma 4 12B Q4_K_M

7.2GB 128k ctx

Runs comfortably

542 tok/s

ⓘ estimated

7 / 748GB

70 / 74 / 72

FIT Runs comfortably

SPEED 542 tok/s

ⓘ estimated

MEMORY 7/748GB

Gemma 4 12B Q8_0

13.0GB 128k ctx

Runs comfortably

300 tok/s

ⓘ estimated

13 / 748GB

70 / 74 / 72

FIT Runs comfortably

SPEED 300 tok/s

ⓘ estimated

MEMORY 13/748GB

Gemma 3 27B Q8_0

28.0GB 128k ctx

Runs comfortably

139 tok/s

ⓘ estimated

28 / 748GB

68 / 71 / 72

FIT Runs comfortably

SPEED 139 tok/s

ⓘ estimated

MEMORY 28/748GB

Gemma 3 27B Q4_K_M

17.0GB 128k ctx

Runs comfortably

230 tok/s

ⓘ estimated

17 / 748GB

68 / 71 / 72

FIT Runs comfortably

SPEED 230 tok/s

ⓘ estimated

MEMORY 17/748GB

Qwen3 8B Q8_0

8.5GB 128k ctx

Runs comfortably

459 tok/s

ⓘ estimated

9 / 748GB

67 / 70 / 65

FIT Runs comfortably

SPEED 459 tok/s

ⓘ estimated

MEMORY 9/748GB

Qwen3 8B Q4_K_M

5.2GB 128k ctx

Runs comfortably

751 tok/s

ⓘ estimated

5 / 748GB

67 / 70 / 65

FIT Runs comfortably

SPEED 751 tok/s

ⓘ estimated

MEMORY 5/748GB

Mixtral 8x7B Instruct Q3_K_M

19.0GB 32k ctx

Runs comfortably

744 tok/s

ⓘ estimated

19 / 748GB

62 / 65 / 68

FIT Runs comfortably

SPEED 744 tok/s

ⓘ estimated

MEMORY 19/748GB

Mixtral 8x7B Instruct Q4_K_M

26.5GB 32k ctx

Runs comfortably

533 tok/s

ⓘ estimated

27 / 748GB

62 / 65 / 68

FIT Runs comfortably

SPEED 533 tok/s

ⓘ estimated

MEMORY 27/748GB

Gemma 3 12B Q8_0

13.5GB 128k ctx

Runs comfortably

289 tok/s

ⓘ estimated

14 / 748GB

62 / 65 / 68

FIT Runs comfortably

SPEED 289 tok/s

ⓘ estimated

MEMORY 14/748GB

Gemma 3 12B Q4_K_M

7.5GB 128k ctx

Runs comfortably

521 tok/s

ⓘ estimated

8 / 748GB

62 / 65 / 68

FIT Runs comfortably

SPEED 521 tok/s

ⓘ estimated

MEMORY 8/748GB

Kolibri 1 BF16

156.2GB 1048k ctx

Runs comfortably

25 tok/s

ⓘ estimated

156 / 748GB

62 / 68 / 54

FIT Runs comfortably

SPEED 25 tok/s

ⓘ estimated

MEMORY 156/748GB

Mistral Nemo 12B Q8_0

13.2GB 128k ctx

Runs comfortably

296 tok/s

ⓘ estimated

13 / 748GB

58 / 60 / 62

FIT Runs comfortably

SPEED 296 tok/s

ⓘ estimated

MEMORY 13/748GB

Mistral Nemo 12B Q4_K_M

7.3GB 128k ctx

Runs comfortably

535 tok/s

ⓘ estimated

7 / 748GB

58 / 60 / 62

FIT Runs comfortably

SPEED 535 tok/s

ⓘ estimated

MEMORY 7/748GB

Qwen3 4B Q4_K_M

2.6GB 32k ctx

Runs comfortably

1501 tok/s

ⓘ estimated

3 / 748GB

58 / 60 / 56

FIT Runs comfortably

SPEED 1501 tok/s

ⓘ estimated

MEMORY 3/748GB

Phi-4 Mini Q8_0

4.1GB 128k ctx

Runs comfortably

952 tok/s

ⓘ estimated

4 / 748GB

55 / 58 / 52

FIT Runs comfortably

SPEED 952 tok/s

ⓘ estimated

MEMORY 4/748GB

Phi-4 Mini Q4_K_M

2.5GB 128k ctx

Runs comfortably

1562 tok/s

ⓘ estimated

3 / 748GB

55 / 58 / 52

FIT Runs comfortably

SPEED 1562 tok/s

ⓘ estimated

MEMORY 3/748GB

Llama 3.1 8B Instruct Q8_0

8.0GB 128k ctx

Runs comfortably

488 tok/s

ⓘ estimated

8 / 748GB

52 / 55 / 62

FIT Runs comfortably

SPEED 488 tok/s

ⓘ estimated

MEMORY 8/748GB

Llama 3.1 8B Instruct FP16

16.0GB 128k ctx

Runs comfortably

244 tok/s

ⓘ estimated

16 / 748GB

52 / 55 / 62

FIT Runs comfortably

SPEED 244 tok/s

ⓘ estimated

MEMORY 16/748GB

Llama 3.1 8B Instruct Q4_K_M

4.5GB 128k ctx

Runs comfortably

867 tok/s

ⓘ estimated

5 / 748GB

52 / 55 / 62

FIT Runs comfortably

SPEED 867 tok/s

ⓘ estimated

MEMORY 5/748GB

Gemma 3 4B Q4_K_M

3.0GB 128k ctx

Runs comfortably

1301 tok/s

ⓘ estimated

3 / 748GB

42 / 45 / 55

FIT Runs comfortably

SPEED 1301 tok/s

ⓘ estimated

MEMORY 3/748GB

Gemma 3 4B Q8_0

5.0GB 128k ctx

Runs comfortably

781 tok/s

ⓘ estimated

5 / 748GB

42 / 45 / 55

FIT Runs comfortably

SPEED 781 tok/s

ⓘ estimated

MEMORY 5/748GB

Llama 3.2 3B Instruct Q4_K_M

2.0GB 128k ctx

Runs comfortably

1952 tok/s

ⓘ estimated

2 / 748GB

35 / 38 / 50

FIT Runs comfortably

SPEED 1952 tok/s

ⓘ estimated

MEMORY 2/748GB

Llama 3.2 3B Instruct Q8_0

3.5GB 128k ctx

Runs comfortably

1115 tok/s

ⓘ estimated

4 / 748GB

35 / 38 / 50

FIT Runs comfortably

SPEED 1115 tok/s

ⓘ estimated

MEMORY 4/748GB

Mellum2.1 12B Thinking Q4_K_M

7.0GB 131k ctx

Runs comfortably

2346 tok/s

ⓘ estimated

7 / 748GB

FIT Runs comfortably

SPEED 2346 tok/s

ⓘ estimated

MEMORY 7/748GB

Mellum2.1 12B Thinking FP16

24.3GB 131k ctx

Runs comfortably

748 tok/s

ⓘ estimated

24 / 748GB

FIT Runs comfortably

SPEED 748 tok/s

ⓘ estimated

MEMORY 24/748GB

Muse Glimmer 30B FP16

60.0GB 128k ctx

Runs comfortably

63 tok/s

ⓘ estimated

60 / 748GB

FIT Runs comfortably

SPEED 63 tok/s

ⓘ estimated

MEMORY 60/748GB

Muse Glimmer 30B Q8_0

33.0GB 128k ctx

Runs comfortably

110 tok/s

ⓘ estimated

33 / 748GB

FIT Runs comfortably

SPEED 110 tok/s

ⓘ estimated

MEMORY 33/748GB

Muse Glimmer 30B Q4_K_M

18.0GB 128k ctx

Runs comfortably

191 tok/s

ⓘ estimated

18 / 748GB

FIT Runs comfortably

SPEED 191 tok/s

ⓘ estimated

MEMORY 18/748GB

Muse Spark Open 816B-A42B UD-Q2_K_XL

311.0GB 1024k ctx

Runs comfortably

244 tok/s

ⓘ estimated

311 / 748GB

FIT Runs comfortably

SPEED 244 tok/s

ⓘ estimated

MEMORY 311/748GB

Muse Spark Open 816B-A42B UD-IQ1_S

176.0GB 1024k ctx

Runs comfortably

431 tok/s

ⓘ estimated

176 / 748GB

FIT Runs comfortably

SPEED 431 tok/s

ⓘ estimated

MEMORY 176/748GB

MiniMax M2.7 UD-IQ1_M Unsloth

60.7GB 200k ctx

Runs comfortably

63 tok/s

ⓘ estimated

61 / 748GB

FIT Runs comfortably

SPEED 63 tok/s

ⓘ estimated

MEMORY 61/748GB

MiniMax M2.7 UD-Q4_K_XL Unsloth

140.8GB 200k ctx

Runs comfortably

28 tok/s

ⓘ estimated

141 / 748GB

FIT Runs comfortably

SPEED 28 tok/s

ⓘ estimated

MEMORY 141/748GB

MiniMax M2.7 UD-Q2_K_XL Unsloth

75.3GB 200k ctx

Runs comfortably

51 tok/s

ⓘ estimated

75 / 748GB

FIT Runs comfortably

SPEED 51 tok/s

ⓘ estimated

MEMORY 75/748GB

MiniMax M3 MXFP4

256.5GB 1048k ctx

Runs comfortably

15 tok/s

ⓘ estimated

257 / 748GB

FIT Runs comfortably

SPEED 15 tok/s

ⓘ estimated

MEMORY 257/748GB

MiniMax M3 UD-Q4_K_XL Unsloth

264.9GB 1048k ctx

Runs comfortably

15 tok/s

ⓘ estimated

265 / 748GB

FIT Runs comfortably

SPEED 15 tok/s

ⓘ estimated

MEMORY 265/748GB

MiniMax M3 UD-IQ1_M Unsloth

128.4GB 1048k ctx

Runs comfortably

30 tok/s

ⓘ estimated

128 / 748GB

FIT Runs comfortably

SPEED 30 tok/s

ⓘ estimated

MEMORY 128/748GB

MiniMax M3 UD-Q2_K_XL Unsloth

143.0GB 1048k ctx

Runs comfortably

27 tok/s

ⓘ estimated

143 / 748GB

FIT Runs comfortably

SPEED 27 tok/s

ⓘ estimated

MEMORY 143/748GB

LFM2.5-24B-A2B FP8

24.0GB 131k ctx

Runs comfortably

1681 tok/s

ⓘ estimated

24 / 748GB

FIT Runs comfortably

SPEED 1681 tok/s

ⓘ estimated

MEMORY 24/748GB

LFM2.5-24B-A2B Q4_K_M

13.7GB 131k ctx

Runs comfortably

2669 tok/s

ⓘ estimated

14 / 748GB

FIT Runs comfortably

SPEED 2669 tok/s

ⓘ estimated

MEMORY 14/748GB

Run GLM-5.3-Flash in the cloud

No hardware? Host it - per-token API or flat subscription

OpenInference per-token

$0.08

/MONTH

Relace per-token

$0.08

/MONTH

StreamLake per-token

$0.11

/MONTH

Wafer per-token

$0.16

/MONTH

GMICloud per-token

$0.18

/MONTH

Sail Research per-token

$0.18

/MONTH

Novita per-token

$0.2

/MONTH

Near AI per-token

$0.28

/MONTH

InferenceNet per-token

$0.29

/MONTH

Phala per-token

$0.31

/MONTH

DigitalOcean per-token

$0.32

/MONTH

Makora per-token

$0.34

/MONTH

Friendli per-token

$0.36

/MONTH

AtlasCloud per-token

$0.36

/MONTH

CoreWeave per-token

$0.36

/MONTH

Venice per-token

$0.36

/MONTH

Z.AI per-token

$0.36

/MONTH

Modal per-token

Serverless GPU inference platform. Serves open-weight models including reason...

$0.36

/MONTH

Crusoe per-token

$0.36

/MONTH

BaseTen per-token

$0.36

/MONTH

Parasail per-token

$0.36

/MONTH

Z.ai per-token

Official API and platform for Zhipu AI's open-source GLM models. GLM-5.2 avai...

pricing may be outdated

$0.36

/MONTH

Together per-token

$0.36

/MONTH

Cohere per-token

$0.36

/MONTH

SiliconFlow per-token

$0.36

/MONTH

DeepInfra per-token

$0.36

/MONTH

PrimeIntellect per-token

$0.36

/MONTH

Morph per-token

$0.43

/MONTH

NextBit per-token

$0.45

/MONTH

Inceptron per-token

$0.47

/MONTH

Fireworks per-token

$0.54

/MONTH

Reka per-token

$0.82

/MONTH

Z.ai Coding Plan Lite subscription

Official API and platform for Zhipu AI's open-source GLM models. GLM-5.2 avai...

$10.0

/MONTH

OpenCode Go Go ($5 first month) subscription

Low-cost subscription for open coding models. $5 first month, then $10/month....

$10.0

/MONTH

Z.ai Coding Plan Pro subscription

Official API and platform for Zhipu AI's open-source GLM models. GLM-5.2 avai...

$30.0

/MONTH

Z.ai Coding Plan Max subscription

Official API and platform for Zhipu AI's open-source GLM models. GLM-5.2 avai...

$80.0

/MONTH

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 →