A
Amazon Bedrock
AGGREGATEDINFERENCE
N/A
Uptime
N/A
Rating
30-Day Uptime
96.7%2026-07-212026-08-19
Inference Latency
Amazon: Nova Micro 1.0366ms TTFT · 175 TPS
Amazon: Nova Micro 1.0913ms TTFT · 54 TPS
Amazon: Nova Lite 1.0344ms TTFT · 104 TPS
Amazon: Nova Lite 1.0948ms TTFT · 55 TPS
OpenAI: gpt-oss-20b341ms TTFT · 285 TPS
OpenAI: gpt-oss-120b482ms TTFT · 140 TPS
OpenAI: gpt-oss-120b428ms TTFT · 328 TPS
OpenAI: GPT-5.6 Luna999ms TTFT · 118 TPS
Anthropic: Claude 3 Haiku590ms TTFT · 31 TPS
Amazon: Nova 2 Lite1035ms TTFT · 92 TPS
Amazon: Nova 2 Lite787ms TTFT · 27 TPS
MoonshotAI: Kimi K2.51003ms TTFT · 60 TPS
Writer: Palmyra X5462ms TTFT · 18 TPS
Meta: Llama 3.1 70B Instruct337ms TTFT · 21 TPS
Amazon: Nova Pro 1.0584ms TTFT · 91 TPS
Amazon: Nova Pro 1.0431ms TTFT · 10 TPS
Z.ai: GLM 52027ms TTFT · 79 TPS
Anthropic: Claude Haiku 4.51054ms TTFT · 50 TPS
Anthropic: Claude Haiku 4.5668ms TTFT · 6 TPS
Anthropic: Claude Sonnet 52939ms TTFT · 47 TPS
Inference Models
| Model | Input $/M | Output $/M | TTFT | TPS |
|---|---|---|---|---|
| Amazon: Nova Micro 1.0 | $0.04 | $0.14 | 366ms | 175 |
| Amazon: Nova Micro 1.0 | $0.04 | $0.14 | 913ms | 54 |
| Amazon: Nova Lite 1.0 | $0.06 | $0.24 | 344ms | 104 |
| Amazon: Nova Lite 1.0 | $0.06 | $0.24 | 948ms | 55 |
| OpenAI: gpt-oss-20b | $0.07 | $0.15 | — | — |
| OpenAI: gpt-oss-20b | $0.07 | $0.15 | 341ms | 285 |
| OpenAI: gpt-oss-120b | $0.15 | $0.60 | 482ms | 140 |
| OpenAI: gpt-oss-120b | $0.15 | $0.60 | 428ms | 328 |
| Qwen: Qwen3 Coder 30B A3B Instruct | $0.15 | $0.60 | — | — |
| OpenAI: GPT-5.6 Luna | $0.22 | $1.32 | 999ms | 118 |
| Anthropic: Claude 3 Haiku | $0.25 | $1.25 | 590ms | 31 |
| Amazon: Nova 2 Lite | $0.30 | $2.50 | 1035ms | 92 |
| Amazon: Nova 2 Lite | $0.30 | $2.50 | 787ms | 27 |
| MoonshotAI: Kimi K2.5 | $0.60 | $3.00 | 1003ms | 60 |
| Writer: Palmyra X5 | $0.60 | $6.00 | 462ms | 18 |
| Meta: Llama 3.1 70B Instruct | $0.72 | $0.72 | 337ms | 21 |
| Amazon: Nova Pro 1.0 | $0.80 | $3.20 | 584ms | 91 |
| Amazon: Nova Pro 1.0 | $0.80 | $3.20 | 431ms | 10 |
| Anthropic: Claude 3.5 Haiku | $0.80 | $4.00 | — | — |
| Anthropic: Claude 3.5 Haiku | $0.80 | $4.00 | — | — |
| Z.ai: GLM 5 | $1.00 | $3.20 | 2027ms | 79 |
| Anthropic: Claude Haiku 4.5 | $1.00 | $5.00 | 1054ms | 50 |
| Anthropic: Claude Haiku 4.5 | $1.10 | $5.50 | — | — |
| Anthropic: Claude Haiku 4.5 | $1.10 | $5.50 | 668ms | 6 |
| Anthropic: Claude Sonnet 5 | $2.00 | $10.00 | 2939ms | 47 |
| Anthropic: Claude Sonnet 5 | $2.20 | $11.00 | 1281ms | 12 |
| OpenAI: GPT-5.6 Terra | $2.20 | $13.20 | 751ms | 58 |
| Amazon: Nova Premier 1.0 | $2.50 | $12.50 | 643ms | 31 |
| OpenAI: GPT-5.4 | $2.75 | $16.50 | — | — |
| Anthropic: Claude Sonnet 4 | $3.00 | $15.00 | — | — |
| Anthropic: Claude Sonnet 4.6 | $3.00 | $15.00 | 1199ms | 56 |
| Anthropic: Claude Sonnet 4.5 | $3.00 | $15.00 | 2034ms | 36 |
| Anthropic: Claude Sonnet 4 | $3.00 | $15.00 | 1054ms | 51 |
| Anthropic: Claude Sonnet 4.5 | $3.30 | $16.50 | 2538ms | 39 |
| Anthropic: Claude Sonnet 4.6 | $3.30 | $16.50 | — | — |
| Anthropic: Claude Sonnet 4.6 | $3.30 | $16.50 | — | — |
| Anthropic: Claude Opus 4.6 | $5.00 | $25.00 | 1433ms | 35 |
| Anthropic: Claude Opus 4.7 | $5.00 | $25.00 | 1637ms | 49 |
| Anthropic: Claude Opus 4.8 | $5.00 | $25.00 | 1811ms | 47 |
| Anthropic: Claude Opus 4.5 | $5.00 | $25.00 | 1664ms | 39 |
| Claude Opus 5 | $5.00 | $25.00 | 3327ms | 61 |
| OpenAI: GPT-5.6 Sol | $5.50 | $33.00 | — | — |
| Claude Opus 5 | $5.50 | $27.50 | 2942ms | 58 |
| OpenAI: GPT-5.5 | $5.50 | $33.00 | 15122ms | 90 |
| Anthropic: Claude Opus 4.7 | $5.50 | $27.50 | — | — |
| Anthropic: Claude Opus 4.8 | $5.50 | $27.50 | 4863ms | 8 |
| Anthropic: Claude Opus 4.5 | $5.50 | $27.50 | 1964ms | 46 |
| Anthropic: Claude Opus 4.8 | $5.50 | $27.50 | 1076ms | 65 |
| Anthropic: Claude 3.5 Sonnet | $6.00 | $30.00 | — | — |
| Anthropic: Claude Fable 5 | $10.00 | $50.00 | — | — |
| Anthropic: Claude Opus 4.1 | $15.00 | $75.00 | 3697ms | 8 |
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.