P
Parasail
AGGREGATEDINFERENCE
N/A
Uptime
N/A
Rating
30-Day Uptime
100%2026-09-192026-10-18
Inference Latency
OpenAI: gpt-oss-20b896ms TTFT · 90 TPS
Mistral: Mistral Nemo319ms TTFT · 112 TPS
Sao10K: Llama 3 8B Lunaris517ms TTFT · 122 TPS
Google: Gemma 3 4B487ms TTFT · 141 TPS
Meta: Llama 3.2 3B Instruct385ms TTFT · 198 TPS
Google: Gemma 3 27B851ms TTFT · 35 TPS
MythoMax 13B401ms TTFT · 96 TPS
Mistral: Mistral Small 3.2 24B357ms TTFT · 50 TPS
Qwen: Qwen3 Next 80B A3B Instruct453ms TTFT · 104 TPS
AllenAI: Olmo 3 7B Instruct187ms TTFT · 44 TPS
OpenAI: gpt-oss-120b464ms TTFT · 156 TPS
Qwen: Qwen3.5-9B421ms TTFT · 89 TPS
ByteDance: UI-TARS 7B 1622ms TTFT · 53 TPS
AllenAI: Olmo 3 7B Think183ms TTFT · 78 TPS
Qwen: Qwen3 Coder Next756ms TTFT · 53 TPS
Google: Gemma 4 26B A4B 1108ms TTFT · 38 TPS
DeepSeek: DeepSeek V4 Flash 0731664ms TTFT · 65 TPS
DeepSeek: DeepSeek V4 Flash 0423665ms TTFT · 53 TPS
Qwen: Qwen3 235B A22B Instruct 2507460ms TTFT · 35 TPS
Qwen: Qwen3.5-35B-A3B639ms TTFT · 181 TPS
Inference Models
| Model | Input $/M | Output $/M | TTFT | TPS |
|---|---|---|---|---|
| OpenAI: gpt-oss-20b | $0.03 | $0.15 | 896ms | 90 |
| Mistral: Mistral Nemo | $0.03 | $0.03 | 319ms | 112 |
| Sao10K: Llama 3 8B Lunaris | $0.04 | $0.05 | 517ms | 122 |
| Google: Gemma 3 4B | $0.04 | $0.05 | 487ms | 141 |
| Meta: Llama 3.2 3B Instruct | $0.05 | $0.33 | 385ms | 198 |
| Google: Gemma 3 27B | $0.08 | $0.45 | 851ms | 35 |
| MythoMax 13B | $0.08 | $0.11 | 401ms | 96 |
| Mistral: Mistral Small 3.2 24B | $0.09 | $0.30 | 357ms | 50 |
| Qwen: Qwen3 Next 80B A3B Instruct | $0.10 | $1.10 | 453ms | 104 |
| AllenAI: Olmo 3 7B Instruct | $0.10 | $0.20 | 187ms | 44 |
| OpenAI: gpt-oss-120b | $0.10 | $0.75 | 464ms | 156 |
| Qwen: Qwen3.5-9B | $0.10 | $0.25 | 421ms | 89 |
| ByteDance: UI-TARS 7B | $0.10 | $0.20 | 1622ms | 53 |
| AllenAI: Olmo 3 7B Think | $0.12 | $0.20 | 183ms | 78 |
| Qwen: Qwen3 Coder Next | $0.12 | $0.80 | 756ms | 53 |
| Google: Gemma 4 26B A4B | $0.13 | $0.40 | 1108ms | 38 |
| DeepSeek: DeepSeek V4 Flash 0731 | $0.14 | $0.28 | 664ms | 65 |
| DeepSeek: DeepSeek V4 Flash 0423 | $0.14 | $0.28 | 665ms | 53 |
| Qwen: Qwen3 235B A22B Instruct 2507 | $0.14 | $0.80 | 460ms | 35 |
| Qwen: Qwen3.5-35B-A3B | $0.15 | $1.00 | 639ms | 181 |
| AllenAI: Olmo 3 32B Think | $0.15 | $0.50 | 263ms | 79 |
| AllenAI: Olmo 3.1 32B Think | $0.15 | $0.50 | 244ms | 81 |
| Z.ai: GLM 5.3 Flash | $0.15 | $0.50 | 867ms | 79 |
| Qwen: Qwen3.6 35B A3B | $0.15 | $1.00 | 616ms | 62 |
| Google: Gemma 4 31B | $0.15 | $0.40 | 1896ms | 19 |
| Z.ai: GLM 5.3 Flash | $0.19 | $0.63 | 389ms | 105 |
| AllenAI: Molmo2 8B | $0.20 | $0.20 | 283ms | 39 |
| Qwen: Qwen3 VL 235B A22B Instruct | $0.21 | $1.90 | 1704ms | 23 |
| Meta: Llama 3.3 70B Instruct | $0.22 | $0.50 | 672ms | 41 |
| Qwen: Qwen3.8 27B | $0.24 | $2.20 | 1152ms | 49 |
| Qwen: Qwen3 VL 8B Instruct | $0.25 | $0.75 | 888ms | 57 |
| TheDrummer: Rocinante 12B | $0.25 | $0.50 | — | — |
| DeepSeek: DeepSeek V4.1 Flash | $0.30 | $1.20 | 406ms | 161 |
| DeepSeek: DeepSeek V4.1 Flash 1017 | $0.30 | $1.20 | 908ms | 106 |
| TheDrummer: Cydonia 24B V4.1 | $0.30 | $0.50 | 649ms | 50 |
| MiniMax: MiniMax M3 | $0.30 | $1.20 | 759ms | 98 |
| Meta: Llama 4 Maverick | $0.35 | $1.00 | 588ms | 54 |
| TheDrummer: UnslopNemo 12B | $0.40 | $0.40 | 487ms | 53 |
| DeepSeek: DeepSeek V4 Pro 0423 | $0.45 | $3.48 | 1423ms | 34 |
| Qwen: Qwen3.5 397B A17B | $0.50 | $3.60 | 432ms | 72 |
| TheDrummer: Skyfall 36B V2 | $0.55 | $0.80 | 443ms | 69 |
| MoonshotAI: Kimi K2.6 | $0.75 | $3.50 | 1696ms | 46 |
| Qwen: Qwen2.5 VL 72B Instruct | $0.80 | $1.00 | 859ms | 34 |
| DeepSeek: DeepSeek V4 Pro 0813 | $1.32 | $3.96 | 1073ms | 47 |
| Z.ai: GLM 5.3 | $1.40 | $4.40 | 736ms | 87 |
| Z.ai: GLM 5.2 | $1.40 | $4.40 | 508ms | 144 |
| MoonshotAI: Kimi K3 | $3.00 | $15.00 | 1343ms | 41 |
| MoonshotAI: Kimi K3 | $3.00 | $15.00 | 859ms | 61 |
Community Reviews
4.5★★★★★(2 reviews)
clouduser42
★★★★★2025-06-15
Reliable service, great API documentation.
mlresearcher
★★★★☆2025-06-10
Good performance but support could be faster.