Parasail
Explore Parasail models on TrustedRouter with current routes, token pricing, policy sources, privacy posture, regional availability, and API support.
Parasailparasail
ZDR
These privacy labels describe Parasail, the upstream model provider. ZDR is a retention policy; verified confidential inference additionally requires attested provider compute and end-to-end encryption.
| Provider | Parasail |
|---|---|
| Routing status | Active |
| Provider website | https://www.parasail.io/ |
| Models | 26 public models |
| Credits routes | 26 |
| Zero data retention | yes All routes |
| Verified confidential inference | Not verified |
| Policy note | Tracked as ZDR for serverless and dedicated inference. Parasail documents no storage or logging of submitted input on those service paths, retention only while generating and delivering output, and no training on input or output. Batch service is excluded from this claim; TrustedRouter does not route Parasail traffic through batch. Policy source |
Measured performance
28 samplesContinuously sampled across Parasail's routed models: p50 TTFT, effective throughput, and success rate. Effective throughput uses provider-reported output tokens over complete request time. Unsupported route and probe-configuration rows are separated from provider downtime. No prompt or output content stored.
| p50 TTFT | 1794 ms |
|---|---|
| Effective throughput | 82 tok/s n=2 |
| Uptime | 89.29% |
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| thedrummer/cydonia-24b-v4.1 | 772 ms | — | 100.00% | — | 1 |
| qwen/qwen2.5-vl-72b-instruct | 851 ms | — | 100.00% | — | 1 |
| deepseek/deepseek-v4-flash | 884 ms | 62 tok/s n=1 | 100.00% | — | 2 |
| meta-llama/llama-3.3-70b-instruct | 1123 ms | — | 100.00% | — | 2 |
| google/gemma-3-27b-it | 1393 ms | — | 100.00% | — | 1 |
| moonshotai/kimi-k2.6 | 1474 ms | — | 66.67% | — | 3 |
| qwen/qwen3-next-80b-a3b-instruct | 1717 ms | — | 100.00% | — | 1 |
| minimax/minimax-m3 | 1794 ms | — | 100.00% | — | 2 |
| z-ai/glm-5.2 | 1992 ms | 103 tok/s n=1 | 100.00% | — | 1 |
| openai/gpt-oss-120b | 2030 ms | — | 100.00% | — | 3 |
| deepseek/deepseek-v4-pro | 2084 ms | — | 100.00% | — | 2 |
| qwen/qwen3-coder-next | 2734 ms | — | 100.00% | — | 1 |
| z-ai/glm-5.3-flash | 3594 ms | — | 100.00% | — | 3 |
| thedrummer/skyfall-36b-v2 | 3920 ms | — | 100.00% | — | 1 |
| openai/gpt-oss-20b | 4039 ms | — | 100.00% | — | 1 |
| deepseek/deepseek-v4-pro-0423 | 6715 ms | — | 100.00% | — | 1 |
| google/gemma-4-26b-a4b-it | — | — | 0.00% | — | 1 |
| qwen/qwen3.5-397b-a17b | — | — | 0.00% | — | 1 |
Parasail performance history · Full provider & model leaderboard.
Models served by Parasail.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Input | Cached input | Output |
|---|---|---|---|---|---|
bytedance/ui-tars-1.5-7bByteDance: UI-TARS 7B |
— | 128,000 | $0.1055/1M | $0.1055/1M | $0.211/1M |
deepseek/deepseek-v4-flashDeepSeek: DeepSeek V4 Flash 0423 |
IQ 116#37 | 1,048,576 | $0.1477/1M | $0.07385/1M | $0.2954/1M |
deepseek/deepseek-v4-proDeepSeek: DeepSeek V4 Pro 0423 |
IQ 114#47 | 1,048,576 | $1.8357/1M | $0.1055/1M | $3.6714/1M |
deepseek/deepseek-v4-pro-0423DeepSeek V4 Pro 0423 |
— | 1,048,576 | $1.8357/1M | $0.1055/1M | $3.6714/1M |
google/gemma-3-27b-itGoogle: Gemma 3 27B |
— | 131,072 | $0.0844/1M | $0.0422/1M | $0.47475/1M |
google/gemma-4-26b-a4b-itGoogle: Gemma 4 26B A4B |
IQ 100#89 | 262,144 | $0.13715/1M | $0.05275/1M | $0.422/1M |
google/gemma-4-31b-itGoogle: Gemma 4 31B |
IQ 103#80 | 262,144 | $0.15825/1M | $0.0633/1M | $0.422/1M |
meta-llama/llama-3.3-70b-instructMeta: Llama 3.3 70B Instruct |
— | 131,072 | $0.2321/1M | $0.11605/1M | $0.5275/1M |
meta-llama/llama-4-maverickMeta: Llama 4 Maverick |
IQ 85#127 | 128,000 | $0.36925/1M | $0.17935/1M | $1.055/1M |
minimax/minimax-m3MiniMax: MiniMax M3 |
IQ 115#45 | 524,288 | $0.3165/1M | $0.0633/1M | $1.266/1M |
moonshotai/kimi-k2.6MoonshotAI: Kimi K2.6 |
IQ 118#35 | 262,144 | $0.79125/1M | $0.1688/1M | $3.6925/1M |
openai/gpt-oss-120bOpenAI: gpt-oss-120b |
IQ 98#99 | 131,072 | $0.1055/1M | $0.058025/1M | $0.79125/1M |
openai/gpt-oss-20bOpenAI: gpt-oss-20b |
IQ 93#113 | 131,072 | $0.0422/1M | $0.0211/1M | $0.211/1M |
qwen/qwen2.5-vl-72b-instructQwen: Qwen2.5 VL 72B Instruct |
— | 128,000 | $0.844/1M | $0.422/1M | $1.055/1M |
qwen/qwen3-coder-nextQwen: Qwen3 Coder Next |
— | 262,144 | $0.1266/1M | $0.07385/1M | $0.844/1M |
qwen/qwen3-next-80b-a3b-instructQwen: Qwen3 Next 80B A3B Instruct |
— | 262,144 | $0.1055/1M | $0.07385/1M | $1.1605/1M |
qwen/qwen3-vl-235b-a22b-instructQwen: Qwen3 VL 235B A22B Instruct |
— | 262,144 | $0.22155/1M | $0.1055/1M | $2.0045/1M |
qwen/qwen3-vl-8b-instructQwen: Qwen3 VL 8B Instruct |
— | 262,144 | $0.26375/1M | $0.1266/1M | $0.79125/1M |
qwen/qwen3.5-35b-a3bQwen: Qwen3.5-35B-A3B |
— | 262,144 | $0.15825/1M | $0.05275/1M | $1.055/1M |
qwen/qwen3.5-397b-a17bQwen: Qwen3.5 397B A17B |
— | 262,144 | $0.5275/1M | $0.3165/1M | $3.798/1M |
qwen/qwen3.6-35b-a3bQwen: Qwen3.6 35B A3B |
IQ 100#94 | 262,144 | $0.15825/1M | $0.05275/1M | $1.055/1M |
thedrummer/cydonia-24b-v4.1TheDrummer: Cydonia 24B V4.1 |
— | 131,072 | $0.3165/1M | $0.15825/1M | $0.5275/1M |
thedrummer/skyfall-36b-v2TheDrummer: Skyfall 36B V2 |
— | 32,768 | $0.58025/1M | $0.26375/1M | $0.844/1M |
z-ai/glm-5.2Z.ai: GLM 5.2 |
IQ 120#28 | 1,048,576 | $1.477/1M | $0.2743/1M | $4.642/1M |
z-ai/glm-5.3Z.ai: GLM 5.3 |
IQ 123#18 | 1,048,576 | $1.477/1M | $0.2743/1M | $4.642/1M |
z-ai/glm-5.3-flashZ.ai: GLM 5.3 Flash |
IQ 116#40 | 1,048,576 | $0.15825/1M | $0.03165/1M | $0.5275/1M |
Questions
Does Parasail have zero data retention?
TrustedRouter records Parasail as supporting provider-level zero data retention based on the policy source linked on this page. This is a provider policy claim, separate from TrustedRouter's content-stateless real-time gateway and from end-to-end confidential compute.
Is Parasail end-to-end encrypted?
TrustedRouter does not currently mark Parasail as end-to-end encrypted at the provider boundary. The TrustedRouter gateway is still attested, but the selected provider normally receives the request in order to run the model. Use trustedrouter/e2e for the stronger route requirement.
Which Parasail models are available through TrustedRouter?
This page currently lists 26 public Parasail models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.