Z.ai: GLM 5.3- Wafer fp4$0.070 / $1.40
- Baidu fp8$0.280 / $0.880
- Morph fp8$0.158 / $1.54
- Reka unknown$0.140 / $2.10
- Sail Research fp8$0.500 / $2.00
- DekaLLM unknown$0.250 / $2.95
- InferenceNet fp4$0.080 / $3.60
- Novita fp8$0.630 / $1.98
- DeepInfra fp4$0.563 / $2.50
- Io Net fp8$0.330 / $3.20
- Tenstorrent fp8$0.600 / $2.50
- SiliconFlow fp8$0.700 / $2.20
- GMICloud fp8$0.700 / $2.20
- Relace unknown$0.070 / $4.38
- Makora fp4$0.140 / $4.40
- AkashML fp8$0.190 / $4.40
- Phala unknown$0.840 / $2.64
- Decart fp4$0.840 / $2.69
- DigitalOcean unknown$0.910 / $2.86
- Inceptron fp4$1.20 / $2.20
- Fireworks unknown$0.980 / $3.08
- PrimeIntellect unknown$0.980 / $3.08
- Alibaba unknown$1.19 / $3.74
- CoreWeave fp8$1.20 / $4.00
- Friendli unknown$1.26 / $3.96
- Nebius fp4$1.40 / $4.40
- BaseTen fp4$1.40 / $4.40
- Mistral nvfp4$1.40 / $4.40
- Venice unknown$1.40 / $4.40
- Together unknown$1.40 / $4.40
- Crusoe fp4$1.40 / $4.40
- Modal unknown$1.40 / $4.40
- Cloudflare unknown$1.40 / $4.40
- AtlasCloud fp8$1.40 / $4.40
- Z.AI fp8$1.40 / $4.40
- Parasail fp8$1.40 / $4.40
- Amazon Bedrock unknown$1.68 / $5.28
- BaseTen fp8$2.10 / $6.60
- ModelRun fp4$2.10 / $6.60
z-ai/glm-5.3 | Wafer fp4 | $0.070 | $1.40 | $0.402 | 39 | 701% |
DeepSeek: DeepSeek V4.1 Flash 1017- Sail Research fp4$0.101 / $0.150
- DeepInfra fp8$0.078 / $0.235
- Morph fp8$0.044 / $0.383
- Wafer mxfp4$0.038 / $0.500
- Relace unknown$0.070 / $0.500
- Decart fp4$0.103 / $0.414
- Reka unknown$0.060 / $0.600
- InferenceNet fp8$0.080 / $0.550
- Ionstream unknown$0.150 / $0.450
- Cloudflare unknown$0.100 / $0.600
- StreamLake fp8$0.150 / $0.600
- DeepSeek unknown$0.150 / $0.600
- Baidu fp8$0.150 / $0.605
- Alibaba unknown$0.165 / $0.660
- ModelRun fp4$0.170 / $0.670
- Cohere unknown$0.195 / $0.660
- DigitalOcean unknown$0.180 / $0.720
- GMICloud fp8$0.180 / $0.720
- Novita fp8$0.195 / $0.780
- AtlasCloud fp8$0.195 / $0.780
- Fireworks unknown$0.210 / $0.840
- Makora fp8$0.150 / $1.20
- Particle AI unknown$0.250 / $1.00
- Modal unknown$0.250 / $1.00
- Phala unknown$0.200 / $1.20
- OpenInference fp4$0.034 / $1.80
- Crusoe fp8$0.290 / $1.20
- SiliconFlow fp8$0.300 / $1.20
- PrimeIntellect unknown$0.300 / $1.20
- Nebius fp8$0.300 / $1.20
- BaseTen unknown$0.300 / $1.20
- Parasail fp8$0.300 / $1.20
- Together unknown$0.300 / $1.20
- Mistral fp8$0.300 / $1.20
- Venice fp8$0.300 / $1.20
deepseek/deepseek-v4.1-flash-1017 | Sail Research fp4 | $0.101 | $0.150 | $0.113 | 35 | 364% |
Z.ai: GLM 5.3 Flash- OpenInference fp4$0.048 / $0.063
- Relace unknown$0.046 / $0.074
- Wafer nvfp4$0.054 / $0.200
- Darkbloom fp4$0.065 / $0.180
- GMICloud fp8$0.075 / $0.250
- Sail Research fp4$0.078 / $0.250
- Novita fp8$0.084 / $0.280
- StreamLake fp8$0.088 / $0.295
- Morph fp8$0.138 / $0.226
- Reka unknown$0.120 / $0.300
- InferenceNet fp4$0.108 / $0.350
- DigitalOcean unknown$0.113 / $0.375
- Near AI fp8$0.080 / $0.500
- CoreWeave nvfp4$0.120 / $0.380
- Phala nvfp4$0.128 / $0.425
- NextBit fp4$0.120 / $0.480
- Makora unknown$0.140 / $0.470
- SiliconFlow fp8$0.150 / $0.500
- Particle AI nvfp4$0.150 / $0.500
- Mistral unknown$0.150 / $0.500
- Together unknown$0.150 / $0.500
- Crusoe fp4$0.150 / $0.500
- AtlasCloud fp8$0.150 / $0.500
- Fireworks unknown$0.150 / $0.500
- Friendli unknown$0.150 / $0.500
- Z.AI fp8$0.150 / $0.500
- DeepInfra fp4$0.150 / $0.500
- Venice unknown$0.150 / $0.500
- Parasail fp4$0.150 / $0.500
- Cohere unknown$0.150 / $0.500
- PrimeIntellect unknown$0.150 / $0.500
- Nebius fp4$0.150 / $0.500
- Modal fp4$0.150 / $0.500
- BaseTen fp8$0.150 / $0.500
- Inceptron fp8$0.170 / $0.600
z-ai/glm-5.3-flash | OpenInference fp4 | $0.048 | $0.063 | $0.052 | 35 | 437% |
DeepSeek: DeepSeek V4.1 Flash- Baidu fp8$0.045 / $0.179
- Decart fp4$0.090 / $0.180
- Morph fp8$0.044 / $0.383
- Sail Research fp4$0.090 / $0.250
- Wafer mxfp4$0.045 / $0.500
- DeepInfra fp8$0.112 / $0.336
- Relace unknown$0.070 / $0.500
- InferenceNet fp8$0.103 / $0.600
- AtlasCloud fp8$0.135 / $0.540
- StreamLake fp8$0.147 / $0.588
- CoreWeave fp8$0.130 / $0.650
- Reka unknown$0.105 / $0.840
- Alibaba unknown$0.165 / $0.660
- GMICloud fp8$0.180 / $0.720
- Novita fp8$0.195 / $0.780
- Fireworks unknown$0.210 / $0.840
- Phala unknown$0.210 / $0.840
- DigitalOcean unknown$0.225 / $0.900
- Mancer 2 fp8$0.200 / $1.00
- OpenInference fp4$0.017 / $1.80
- Makora fp8$0.260 / $1.17
- Crusoe fp8$0.290 / $1.20
- Venice fp8$0.300 / $1.20
- Together unknown$0.300 / $1.20
- BaseTen unknown$0.300 / $1.20
- Cohere unknown$0.300 / $1.20
- PrimeIntellect unknown$0.300 / $1.20
- Nebius fp8$0.300 / $1.20
- Google unknown$0.300 / $1.20
- SiliconFlow fp8$0.300 / $1.20
- Modal unknown$0.300 / $1.20
- Parasail fp8$0.300 / $1.20
deepseek/deepseek-v4.1-flash | Baidu fp8 | $0.045 | $0.179 | $0.078 | 32 | 571% |
Z.ai: GLM 5.2- Morph fp8$0.158 / $1.54
- DeepInfra fp4$0.450 / $1.44
- StreamLake fp8$0.612 / $1.92
- Ambient fp8$0.600 / $2.00
- Baidu fp8$0.619 / $1.94
- Novita fp8$0.650 / $2.04
- Baidu fp4$0.619 / $2.17
- CoreWeave fp4$0.620 / $2.42
- SiliconFlow fp8$0.700 / $2.20
- DigitalOcean unknown$0.700 / $2.20
- AtlasCloud fp8$0.938 / $2.95
- Alibaba fp8$0.966 / $3.04
- Wafer fp4$0.006 / $7.00
- Relace unknown$0.006 / $7.91
- Inceptron fp4$1.20 / $4.40
- Cloudflare unknown$0.190 / $8.00
- Nebius fp4$1.40 / $4.40
- Mistral nvfp4$1.40 / $4.40
- BaseTen fp8$1.40 / $4.40
- Together unknown$1.40 / $4.40
- Venice fp8$1.40 / $4.40
- GMICloud fp8$1.40 / $4.40
- Parasail fp4$1.40 / $4.40
- Friendli unknown$1.40 / $4.40
- Z.AI fp8$1.40 / $4.40
- Mistral unknown$1.54 / $4.84
- Decart fp4$1.20 / $8.00
z-ai/glm-5.2 | Morph fp8 | $0.158 | $1.54 | $0.504 | 27 | 475% |
DeepSeek: DeepSeek V4 Flash 0731- Relace fp4$0.051 / $0.149
- Baidu fp8$0.054 / $0.161
- DeepInfra fp8$0.060 / $0.180
- Wafer nvfp4$0.100 / $0.230
- Reka unknown$0.100 / $0.280
- DigitalOcean unknown$0.119 / $0.238
- BaseTen fp8$0.130 / $0.260
- BaseTen fp32$0.130 / $0.260
- Venice unknown$0.130 / $0.260
- CoreWeave fp8$0.130 / $0.280
- Parasail fp8$0.140 / $0.280
- Cohere unknown$0.140 / $0.280
- Sail Research fp4$0.150 / $0.300
- StreamLake fp8$0.132 / $0.396
- Inceptron fp4$0.130 / $0.409
- Alibaba unknown$0.176 / $0.528
- Mancer 2 fp8$0.200 / $0.600
- SiliconFlow fp8$0.220 / $0.660
- GMICloud fp8$0.286 / $0.858
- Phala unknown$0.308 / $0.924
- Novita fp8$0.409 / $1.23
- Cloudflare unknown$0.440 / $1.32
- AtlasCloud fp4$0.440 / $1.32
deepseek/deepseek-v4-flash-0731 | Relace fp4 | $0.051 | $0.149 | $0.075 | 23 | 776% |
DeepSeek: DeepSeek V4 Pro 0813- Sail Research fp4$0.450 / $0.800
- Alibaba unknown$0.581 / $1.74
- Baidu fp8$0.659 / $1.98
- StreamLake unknown$0.660 / $1.98
- DeepSeek unknown$0.660 / $1.98
- Ionstream unknown$0.370 / $2.93
- Wafer mxfp4$0.115 / $4.30
- Phala unknown$0.957 / $2.88
- Novita fp8$0.990 / $2.97
- GMICloud fp8$1.06 / $3.17
- DeepInfra fp8$1.30 / $2.60
- DekaLLM unknown$1.08 / $3.80
- CoreWeave fp8$1.31 / $3.96
- Nebius fp8$1.32 / $3.96
- Cloudflare unknown$1.32 / $3.96
- SiliconFlow fp8$1.32 / $3.96
- DigitalOcean unknown$1.32 / $3.96
- Together unknown$1.32 / $3.96
- Parasail fp8$1.32 / $3.96
- AtlasCloud fp8$1.32 / $3.96
- BaseTen fp4$1.32 / $3.96
- Relace fp4$0.103 / $8.00
- Venice unknown$1.65 / $4.95
deepseek/deepseek-v4-pro-0813 | Sail Research fp4 | $0.450 | $0.800 | $0.538 | 23 | 361% |
MoonshotAI: Kimi K3- Relace fp4$0.182 / $9.34
- InferenceNet fp4$0.239 / $10.80
- Sail Research fp4$0.230 / $11.00
- Decart mxfp4$1.65 / $8.25
- Chutes mxfp4$1.85 / $9.50
- Wafer mxfp4$0.192 / $15.00
- Makora fp4$0.500 / $15.00
- Fireworks unknown$2.10 / $10.50
- AkashML fp4$1.20 / $14.00
- Morph fp8$1.28 / $14.63
- Phala unknown$2.55 / $12.75
- DigitalOcean unknown$2.55 / $12.95
- DeepInfra mxfp4$2.85 / $14.25
- Modal mxfp4$3.00 / $15.00
- ParetoAlpha unknown$3.00 / $15.00
- Amazon Bedrock unknown$3.00 / $15.00
- Parasail fp4$3.00 / $15.00
- Together unknown$3.00 / $15.00
- BaseTen fp8$3.00 / $15.00
- Moonshot AI unknown$3.00 / $15.00
- Alibaba unknown$3.45 / $17.25
moonshotai/kimi-k3 | Relace fp4 | $0.182 | $9.34 | $2.47 | 21 | 179% |
Qwen: Qwen3.8 27B- Chutes fp8$0.039 / $1.25
- Near AI fp8$0.040 / $1.35
- Tenstorrent fp4$0.150 / $1.50
- Ionstream fp8$0.040 / $1.95
- Darkbloom fp4$0.035 / $2.20
- Reka fp8$0.040 / $2.20
- DeepInfra bf16$0.150 / $1.88
- Modal unknown$0.040 / $2.30
- Wafer fp8$0.060 / $2.30
- DekaLLM unknown$0.040 / $2.50
- AkashML fp8$0.225 / $1.98
- Parasail fp8$0.240 / $2.20
- Mancer 2 fp8$0.180 / $2.50
- ModelRun fp4$0.200 / $2.50
- Phala unknown$0.200 / $2.50
- Alibaba unknown$0.425 / $2.55
- CoreWeave fp8$0.400 / $3.00
- Novita unknown$0.420 / $3.00
- Cerebras fp16$0.990 / $1.49
- Venice fp8$0.450 / $3.20
- Cloudflare unknown$0.450 / $3.20
qwen/qwen3.8-27b | Chutes fp8 | $0.039 | $1.25 | $0.342 | 21 | 233% |
DeepSeek: DeepSeek V4 Flash 0423- Baidu fp8$0.050 / $0.100
- StreamLake fp8$0.070 / $0.140
- DeepInfra fp8$0.090 / $0.180
- GMICloud fp8$0.091 / $0.182
- Venice unknown$0.097 / $0.193
- DigitalOcean unknown$0.098 / $0.196
- Fireworks unknown$0.105 / $0.210
- Relace fp4$0.047 / $0.450
- SiliconFlow fp8$0.130 / $0.280
- Crusoe fp8$0.140 / $0.280
- AtlasCloud fp4$0.140 / $0.280
- Parasail fp8$0.140 / $0.280
- Alibaba fp8$0.150 / $0.300
- Mancer 2 fp8$0.190 / $0.500
- Azure unknown$0.210 / $0.560
- Wafer mxfp4$0.600 / $0.170
- Cloudflare unknown$0.440 / $1.32
deepseek/deepseek-v4-flash | Baidu fp8 | $0.050 | $0.100 | $0.063 | 17 | 956% |
MoonshotAI: Kimi K2.6- Baidu fp4$0.390 / $1.64
- Chutes int4$0.390 / $1.80
- DigitalOcean unknown$0.570 / $2.40
- StreamLake fp8$0.599 / $2.52
- CoreWeave fp4$0.650 / $3.41
- Inceptron int4$0.624 / $3.50
- Crusoe bf16$0.700 / $3.50
- SiliconFlow fp8$0.770 / $3.40
- Venice int4$0.750 / $3.50
- Parasail int4$0.750 / $3.50
- DeepInfra fp4$0.750 / $3.50
- Novita unknown$0.800 / $3.40
- GMICloud fp8$0.855 / $3.60
- Cloudflare unknown$0.950 / $4.00
- AtlasCloud int4$0.950 / $4.00
- Moonshot AI unknown$0.950 / $4.00
- Phala unknown$1.09 / $4.60
moonshotai/kimi-k2.6 | Baidu fp4 | $0.390 | $1.64 | $0.702 | 17 | 180% |
Google: Gemma 4 26B A4B - Darkbloom unknown$0.042 / $0.220
- AkashML fp4$0.070 / $0.240
- Makora unknown$0.070 / $0.240
- DekaLLM bf16$0.060 / $0.330
- DeepInfra fp8$0.070 / $0.340
- NextBit bf16$0.090 / $0.300
- Io Net bf16$0.090 / $0.310
- Cloudflare unknown$0.100 / $0.300
- PrimeIntellect bf16$0.100 / $0.300
- CoreWeave bf16$0.100 / $0.300
- Parasail bf16$0.130 / $0.400
- Venice bf16$0.130 / $0.400
- Novita bf16$0.130 / $0.400
- SiliconFlow fp8$0.140 / $0.400
- Google unknown$0.150 / $0.600
google/gemma-4-26b-a4b-it | Darkbloom unknown | $0.042 | $0.220 | $0.086 | 15 | 204% |
DeepSeek: DeepSeek V4 Pro 0423- Relace fp4$0.900 / $1.50
- StreamLake fp8$0.887 / $1.77
- Baidu fp8$0.889 / $1.78
- GMICloud fp8$0.957 / $1.91
- DigitalOcean unknown$1.04 / $2.09
- Cloudflare unknown$1.15 / $2.55
- SiliconFlow fp8$1.00 / $3.13
- DeepInfra fp8$1.30 / $2.60
- Alibaba fp8$1.42 / $2.83
- Novita fp8$1.60 / $3.20
- Venice unknown$1.65 / $3.30
- AtlasCloud fp4$1.68 / $3.38
- Crusoe fp8$1.74 / $3.48
- Azure unknown$1.91 / $3.83
deepseek/deepseek-v4-pro | Relace fp4 | $0.900 | $1.50 | $1.05 | 14 | 128% |
Google: Gemma 4 31B- CoreWeave fp4$0.100 / $0.340
- Venice fp4$0.120 / $0.360
- Chutes fp4$0.120 / $0.370
- DeepInfra fp8$0.160 / $0.320
- Crusoe bf16$0.140 / $0.400
- Novita bf16$0.140 / $0.400
- Friendli unknown$0.140 / $0.400
- DekaLLM unknown$0.140 / $0.400
- Parasail fp8$0.150 / $0.400
- DeepInfra fp4$0.135 / $0.510
- Io Net unknown$0.361 / $1.09
- SambaNova unknown$0.380 / $1.15
- SiliconFlow fp8$0.750 / $1.00
- ModelRun fp4$0.750 / $1.00
google/gemma-4-31b-it | CoreWeave fp4 | $0.100 | $0.340 | $0.160 | 14 | 408% |
Z.ai: GLM 5.1- Chutes fp8$0.900 / $2.90
- StreamLake fp8$0.966 / $3.04
- SiliconFlow fp8$1.19 / $3.74
- AtlasCloud fp8$1.26 / $3.96
- Phala unknown$1.21 / $4.20
- Alibaba fp8$1.33 / $4.18
- Novita fp8$1.38 / $4.40
- Baidu fp8$1.40 / $4.40
- GMICloud fp8$1.40 / $4.40
- Friendli unknown$1.40 / $4.40
- Z.AI fp8$1.40 / $4.40
- Nebius fp8$1.40 / $4.40
- Venice fp8$1.40 / $4.40
z-ai/glm-5.1 | Chutes fp8 | $0.900 | $2.90 | $1.40 | 13 | 54% |
MiniMax: MiniMax M3- DeepInfra fp8$0.224 / $0.880
- CoreWeave fp4$0.230 / $0.960
- GMICloud fp8$0.240 / $0.960
- StreamLake fp8$0.300 / $1.20
- Venice fp8$0.300 / $1.20
- Parasail fp8$0.300 / $1.20
- Together unknown$0.300 / $1.20
- AtlasCloud fp8$0.300 / $1.20
- Novita fp8$0.300 / $1.20
- Minimax unknown$0.300 / $1.20
- ModelRun fp8$0.300 / $1.30
- Mara unknown$0.600 / $2.40
minimax/minimax-m3 | DeepInfra fp8 | $0.224 | $0.880 | $0.388 | 12 | 171% |
MiniMax: MiniMax M3.1 Flash- StreamLake fp8$0.135 / $0.540
- GMICloud fp8$0.142 / $0.570
- InferenceNet fp4$0.150 / $0.600
- SambaNova unknown$0.150 / $0.600
- Fireworks unknown$0.150 / $0.600
- ModelRun fp4$0.150 / $0.600
- Together unknown$0.150 / $0.600
- DigitalOcean unknown$0.150 / $0.600
- Novita fp16$0.150 / $0.600
- Nebius fp4$0.150 / $0.600
- Minimax unknown$0.150 / $0.600
- Parasail fp8$0.150 / $0.600
minimax/minimax-m3.1-flash | StreamLake fp8 | $0.135 | $0.540 | $0.236 | 12 | 11% |
MoonshotAI: Kimi K2.7 Code- Inceptron int4$0.560 / $3.10
- CoreWeave int4$0.600 / $3.00
- StreamLake unknown$0.713 / $3.00
- Venice int4$0.750 / $3.50
- SiliconFlow fp8$0.859 / $3.80
- Novita int4$0.912 / $3.84
- Alibaba fp8$0.950 / $4.00
- Cloudflare unknown$0.950 / $4.00
- Moonshot AI unknown$0.950 / $4.00
- GMICloud fp8$0.950 / $4.00
- Nebius fp4$0.950 / $4.00
moonshotai/kimi-k2.7-code | Inceptron int4 | $0.560 | $3.10 | $1.20 | 11 | 43% |
Qwen: Qwen3.5 397B A17B- Alibaba unknown$0.390 / $2.34
- DeepInfra fp8$0.450 / $3.00
- Parasail fp8$0.500 / $3.60
- Phala unknown$0.550 / $3.50
- AtlasCloud fp8$0.550 / $3.50
- DigitalOcean unknown$0.550 / $3.50
- Novita unknown$0.600 / $3.60
- GMICloud fp8$0.600 / $3.60
- StreamLake unknown$0.600 / $3.60
- Venice unknown$0.750 / $4.50
qwen/qwen3.5-397b-a17b | Alibaba unknown | $0.390 | $2.34 | $0.877 | 10 | 92% |
Qwen: Qwen3.6 35B A3B- Darkbloom fp4$0.050 / $0.700
- AkashML fp8$0.100 / $0.900
- Parasail fp8$0.100 / $0.900
- DeepInfra fp8$0.100 / $0.950
- CoreWeave fp8$0.100 / $0.950
- DekaLLM unknown$0.100 / $1.00
- Venice fp8$0.100 / $1.00
- Phala unknown$0.200 / $1.27
- AtlasCloud fp8$0.250 / $1.25
- SiliconFlow fp8$0.240 / $1.80
qwen/qwen3.6-35b-a3b | Darkbloom fp4 | $0.050 | $0.700 | $0.212 | 10 | 197% |
DeepSeek: DeepSeek V3.2- GMICloud fp8$0.209 / $0.310
- AtlasCloud fp8$0.260 / $0.380
- DeepInfra fp4$0.260 / $0.380
- Baidu fp8$0.280 / $0.420
- SiliconFlow fp8$0.259 / $0.800
- DigitalOcean unknown$0.300 / $0.960
- Phala unknown$1.00 / $1.00
- Mara unknown$3.00 / $4.50
- SambaNova unknown$3.00 / $4.50
deepseek/deepseek-v3.2 | GMICloud fp8 | $0.209 | $0.310 | $0.234 | 9 | 1342% |
NVIDIA: Nemotron 3.5 Super VL- Darkbloom nvfp4$0.060 / $0.350
- Crusoe bf16$0.100 / $0.400
- Together unknown$0.150 / $0.500
- Io Net nvfp4$0.150 / $0.500
- CoreWeave nvfp4$0.150 / $0.500
- DeepInfra fp4$0.150 / $0.500
- BaseTen fp4$0.200 / $0.500
- Venice fp4$0.200 / $0.500
- Nebius bf16$0.300 / $0.900
nvidia/nemotron-3.5-super-vl | Darkbloom nvfp4 | $0.060 | $0.350 | $0.133 | 9 | 240% |
Z.ai: GLM 5- GMICloud fp8$0.600 / $1.92
- StreamLake fp8$0.600 / $1.92
- Baidu fp8$0.700 / $2.24
- SiliconFlow fp8$0.950 / $2.55
- Z.AI fp8$1.00 / $3.20
- Amazon Bedrock unknown$1.00 / $3.20
- Venice fp8$1.00 / $3.20
- Novita fp8$1.00 / $3.20
- Venice unknown$1.20 / $4.00
z-ai/glm-5 | GMICloud fp8 | $0.600 | $1.92 | $0.930 | 9 | 104% |
MiniMax: MiniMax M2.5- Venice unknown$0.270 / $0.950
- StreamLake unknown$0.270 / $1.08
- AtlasCloud fp8$0.295 / $1.20
- GMICloud fp8$0.300 / $1.20
- Minimax unknown$0.300 / $1.20
- Novita fp8$0.300 / $1.20
- DigitalOcean unknown$0.300 / $1.20
minimax/minimax-m2.5 | Venice unknown | $0.270 | $0.950 | $0.440 | 7 | 19% |
Qwen: Qwen3.5-35B-A3B- Darkbloom fp4$0.065 / $0.750
- DeepInfra fp8$0.140 / $1.00
- Parasail fp8$0.150 / $1.00
- Venice unknown$0.150 / $1.00
- Alibaba unknown$0.163 / $1.30
- AtlasCloud fp8$0.225 / $1.80
- SiliconFlow fp8$0.240 / $1.80
qwen/qwen3.5-35b-a3b | Darkbloom fp4 | $0.065 | $0.750 | $0.236 | 7 | 167% |
Qwen: Qwen3.6 27B- Chutes fp8$0.250 / $2.00
- Reka unknown$0.200 / $3.00
- Alibaba unknown$0.450 / $2.70
- SiliconFlow fp8$0.300 / $3.20
- DeepInfra fp8$0.320 / $3.20
- Phala unknown$0.320 / $3.25
- Venice fp8$0.325 / $3.25
qwen/qwen3.6-27b | Chutes fp8 | $0.250 | $2.00 | $0.688 | 7 | 54% |
Qwen: Qwen3.5-27B- Alibaba unknown$0.195 / $1.56
- SiliconFlow fp8$0.250 / $2.00
- AtlasCloud fp8$0.270 / $2.16
- Novita bf16$0.300 / $2.40
- Phala unknown$0.300 / $2.40
- DeepInfra fp8$0.260 / $2.60
qwen/qwen3.5-27b | Alibaba unknown | $0.195 | $1.56 | $0.536 | 6 | 58% |
Qwen: Qwen3.5-9B- Darkbloom fp4$0.080 / $0.130
- DeepInfra bf16$0.100 / $0.150
- SiliconFlow fp8$0.100 / $0.150
- Parasail bf16$0.100 / $0.250
- PrimeIntellect bf16$0.170 / $0.250
- Together unknown$0.170 / $0.250
qwen/qwen3.5-9b | Darkbloom fp4 | $0.080 | $0.130 | $0.092 | 6 | 105% |
MiniMax: MiniMax M2.7- GMICloud fp8$0.210 / $0.840
- Novita fp8$0.270 / $1.08
- Minimax unknown$0.300 / $1.20
- AtlasCloud fp8$0.300 / $1.20
- Groq unknown$0.600 / $1.80
minimax/minimax-m2.7 | GMICloud fp8 | $0.210 | $0.840 | $0.367 | 5 | 145% |
MoonshotAI: Kimi K2.5- AtlasCloud int4$0.490 / $2.50
- SiliconFlow int4$0.500 / $2.50
- Novita unknown$0.570 / $2.85
- Amazon Bedrock unknown$0.600 / $3.00
- Venice unknown$0.532 / $3.33
moonshotai/kimi-k2.5 | AtlasCloud int4 | $0.490 | $2.50 | $0.992 | 5 | 24% |
Qwen: Qwen3.8 2.4T A95B- Alibaba unknown$2.00 / $6.00
- Novita unknown$2.00 / $6.00
- SiliconFlow fp8$2.00 / $6.00
- Together unknown$2.00 / $6.00
- Venice unknown$2.00 / $6.00
qwen/qwen3.8-2.4t-a95b | Alibaba unknown | $2.00 | $6.00 | $3.00 | 5 | 0% |
Z.ai: GLM 4.7- DeepInfra fp4$0.320 / $1.40
- Venice fp4$0.400 / $1.93
- Novita fp8$0.540 / $1.98
- Z.AI fp4$0.600 / $2.20
- Mancer 2 fp4$0.700 / $2.50
z-ai/glm-4.7 | DeepInfra fp4 | $0.320 | $1.40 | $0.590 | 5 | 95% |
NVIDIA: Nemotron 3.5 Lightning- Darkbloom int4$0.025 / $0.120
- Io Net unknown$0.047 / $0.134
- DeepInfra bf16$0.060 / $0.160
- CoreWeave bf16$0.070 / $0.200
nvidia/nemotron-3.5-lightning | Darkbloom int4 | $0.025 | $0.120 | $0.049 | 4 | 110% |
Qwen: Qwen3.5-122B-A10B- Alibaba unknown$0.260 / $2.08
- SiliconFlow fp8$0.260 / $2.08
- AtlasCloud fp8$0.300 / $2.40
- Novita bf16$0.400 / $3.20
qwen/qwen3.5-122b-a10b | Alibaba unknown | $0.260 | $2.08 | $0.715 | 4 | 54% |
DeepSeek: DeepSeek V4 Flash Vision Exp- DeepInfra fp8$0.216 / $0.647
- GMICloud fp8$0.440 / $1.32
- SiliconFlow fp8$0.440 / $1.32
deepseek/deepseek-v4-flash-vision-exp | DeepInfra fp8 | $0.216 | $0.647 | $0.323 | 3 | 104% |
NVIDIA: Nemotron 3 Super- DeepInfra bf16$0.085 / $0.400
- DekaLLM fp8$0.080 / $0.450
- Crusoe fp8$0.300 / $2.40
nvidia/nemotron-3-super-120b-a12b | DeepInfra bf16 | $0.085 | $0.400 | $0.164 | 3 | 404% |
NVIDIA: Nemotron 3 Ultra- DeepInfra fp4$0.400 / $1.76
- BaseTen fp4$0.600 / $2.40
- Venice fp8$0.625 / $3.13
nvidia/nemotron-3-ultra-550b-a55b | DeepInfra fp4 | $0.400 | $1.76 | $0.740 | 3 | 69% |
Qwen: Qwen3 VL 30B A3B Instruct- DeepInfra fp8$0.150 / $0.600
- Novita bf16$0.200 / $0.700
- SiliconFlow fp8$0.290 / $1.00
qwen/qwen3-vl-30b-a3b-instruct | DeepInfra fp8 | $0.150 | $0.600 | $0.262 | 3 | 78% |
DeepSeek: DeepSeek V4.1 Flash (batch)- Sail Research fp4$0.050 / $0.150
- DeepInfra fp8$0.112 / $0.336
deepseek/deepseek-v4.1-flash:batch | Sail Research fp4 | $0.050 | $0.150 | $0.075 | 2 | 124% |
NVIDIA: Nemotron 3 Nano 30B A3B- Novita fp4$0.050 / $0.200
- Nebius fp8$0.060 / $0.240
nvidia/nemotron-3-nano-30b-a3b | Novita fp4 | $0.050 | $0.200 | $0.088 | 2 | 20% |
Z.ai: GLM 4.6V- Novita bf16$0.300 / $0.900
- Z.AI fp8$0.300 / $0.900
z-ai/glm-4.6v | Novita bf16 | $0.300 | $0.900 | $0.450 | 2 | 0% |
Z.ai: GLM 4.7 Flash- Cloudflare unknown$0.060 / $0.400
- Novita bf16$0.070 / $0.400
z-ai/glm-4.7-flash | Cloudflare unknown | $0.060 | $0.400 | $0.145 | 2 | 5% |
Z.ai: GLM 5.3 Flash (batch)- Sail Research fp4$0.050 / $0.180
- DeepInfra fp4$0.120 / $0.400
z-ai/glm-5.3-flash:batch | Sail Research fp4 | $0.050 | $0.180 | $0.083 | 2 | 130% |
MiniMax: MiniMax M2- Minimax fp8$0.255 / $1.02
minimax/minimax-m2 | Minimax fp8 | $0.255 | $1.02 | $0.446 | 1 | 0% |
MiniMax: MiniMax M2.1- Minimax unknown$0.300 / $1.20
minimax/minimax-m2.1 | Minimax unknown | $0.300 | $1.20 | $0.525 | 1 | 0% |
Mistral: Devstral 2 2512- Mistral unknown$0.400 / $2.00
mistralai/devstral-2512 | Mistral unknown | $0.400 | $2.00 | $0.800 | 1 | 0% |
Mistral: Ministral 3 14B 2512- Mistral unknown$0.200 / $0.200
mistralai/ministral-14b-2512 | Mistral unknown | $0.200 | $0.200 | $0.200 | 1 | 0% |
Mistral: Ministral 3 3B 2512- Mistral unknown$0.100 / $0.100
mistralai/ministral-3b-2512 | Mistral unknown | $0.100 | $0.100 | $0.100 | 1 | 0% |
Mistral: Ministral 3 8B 2512- Mistral unknown$0.150 / $0.150
mistralai/ministral-8b-2512 | Mistral unknown | $0.150 | $0.150 | $0.150 | 1 | 0% |
Mistral: Mistral Large 4- Mistral unknown$1.36 / $4.18
mistralai/mistral-large-4-0 | Mistral unknown | $1.36 | $4.18 | $2.06 | 1 | 0% |
Mistral: Mistral Small 4- Mistral unknown$0.150 / $0.600
mistralai/mistral-small-2603 | Mistral unknown | $0.150 | $0.600 | $0.262 | 1 | 0% |
Mistral: Voxtral Small 24B 2507- Mistral unknown$0.100 / $0.300
mistralai/voxtral-small-24b-2507 | Mistral unknown | $0.100 | $0.300 | $0.150 | 1 | 0% |
MoonshotAI: Kimi K2 Thinking- Novita bf16$0.600 / $2.50
moonshotai/kimi-k2-thinking | Novita bf16 | $0.600 | $2.50 | $1.07 | 1 | 0% |
MoonshotAI: Kimi K3 (batch)- DeepInfra bf16$2.28 / $11.40
moonshotai/kimi-k3:batch | DeepInfra bf16 | $2.28 | $11.40 | $4.56 | 1 | 0% |
NVIDIA: Nemotron 3.5 Content Safety- DeepInfra bf16$0.200 / $0.200
nvidia/nemotron-3.5-content-safety | DeepInfra bf16 | $0.200 | $0.200 | $0.200 | 1 | 0% |
OpenAI: gpt-oss-safeguard-20b- Groq unknown$0.075 / $0.300
openai/gpt-oss-safeguard-20b | Groq unknown | $0.075 | $0.300 | $0.131 | 1 | 0% |
Qwen: Qwen3 Coder Next- Parasail bf16$0.120 / $0.800
qwen/qwen3-coder-next | Parasail bf16 | $0.120 | $0.800 | $0.290 | 1 | 0% |
Qwen: Qwen3 VL 30B A3B Thinking- SiliconFlow fp8$0.290 / $1.00
qwen/qwen3-vl-30b-a3b-thinking | SiliconFlow fp8 | $0.290 | $1.00 | $0.467 | 1 | 0% |
Qwen: Qwen3 VL 8B Instruct- Parasail bf16$0.250 / $0.750
qwen/qwen3-vl-8b-instruct | Parasail bf16 | $0.250 | $0.750 | $0.375 | 1 | 0% |
Qwen: Qwen3.8 Flash- Alibaba unknown$0.150 / $0.470
qwen/qwen3.8-flash | Alibaba unknown | $0.150 | $0.470 | $0.230 | 1 | 0% |
Z.ai: GLM 5.3 (batch)- DeepInfra fp4$0.450 / $2.00
z-ai/glm-5.3:batch | DeepInfra fp4 | $0.450 | $2.00 | $0.838 | 1 | 0% |