IO Intelligence
Explore IO Intelligence models on TrustedRouter with current routes, token pricing, policy sources, privacy posture, regional availability, and API support.
IO Intelligenceio-net
IO Intelligence is io.net's OpenAI-compatible inference API for open-weight models, with streaming chat completions and an authenticated model and pricing catalog.
These privacy labels describe IO Intelligence, the upstream model provider. Confidential requires all three: verified provider compute, end-to-end encryption and explicit zero data retention. Account and billing metadata are outside the content-retention claim.
| Provider | IO Intelligence |
|---|---|
| API operator country | US Legal home of the API operator, not an inference-location guarantee. |
| Headquarters | West Hollywood, California, United States (operating address) Provider-submitted operating address; not independently verified as headquarters. Source; reviewed 2026-09-28. |
| Legal entity | io.net, Inc.; Delaware corporation, USA |
| Brand | io.net |
| Registered address | 1209 N Orange Street, Wilmington, DE 19801, USA |
| Operating address | 2041 N Alta Vista Ave, West Hollywood, CA 90046, USA |
| EIN | 99-2468828 |
| Company registration and DUNS | Available on request; identifiers not supplied. |
| CEO | Gaurav Sharma |
| Ownership | Privately held |
| Compliance | SOC 2 reported by the provider and listed in its trust center; audit report not reviewed. |
| DPA and subprocessors | Available on request; documents not yet reviewed. |
| API | Bearer-authenticated, OpenAI-compatible streaming chat completions. |
| Pricing | Input, output and cached-input token prices refresh hourly from the authenticated model catalog. Only priced, admitted routes are listed. |
| Rate limits | Provider reports HTTP 429 for rate limits and 503 for overload. Partner limit exemptions require account-specific onboarding confirmation. |
| Provider checker | Report pending; this submission is not a Provider Reliability Contract v2 conformance result. |
| Information reviewed | September 28, 2026. Company and inference declarations supplied by io.net; public policy and trust-center evidence reviewed separately. |
| Routing status | Active |
| Provider website | https://io.net |
| Models | 37 public models |
| Credits routes | 37 |
| Zero data retention | not claimed |
| Verified confidential inference | Not verified |
| Policy note | In its September 28, 2026 marketplace submission, io.net states that inference prompts and completions are processed only in memory, are not stored or logged, and are not used for training. Content-free operational metadata is retained for billing, capacity planning and abuse prevention. This is a provider declaration: the reviewed public privacy policy does not establish inference-specific zero-retention terms, and the applicable DPA has not been reviewed. Routes remain Standard pending that evidence. Confidential compute is offered on request, but this integration has no verified confidential-compute or provider-side E2EE guarantee. Policy source |
Documentation and policy sources
Inference locations
Provider API declarations are also available on each model's endpoint API. Direct chat and Responses replies separate advertised availability from serving location in usage.inference_location; the actual serving region remains unknown when not reported.
Where the upstream GPUs execute the model, not the provider's headquarters, API ingress, or data storage. ZDR and confidential inference do not by themselves restrict geography.
- Declared countries or regions
- United States; Canada
io.net declares US and Canadian serving locations and data residency in its September 28, 2026 marketplace submission. This is provider-wide information, not an independently verified per-model or per-request location. - Can locations change?
- Placement, cross-country failover and future location changes require provider confirmation.
- Provider region pinning
- No per-request country or region pin verified for the shared inference API.
- Pinning through TrustedRouter
- Not supported by the current TrustedRouter integration.
- Infrastructure declaration
- United States and Canada, provider-declared; no US-only or fixed-country guarantee on this route.
- Evidence
- Provider-submitted marketplace information; linked public policy is not an inference-location guarantee. Reviewed .
Provider selection alone is not a region pin. TrustedRouter's US provider filter selects company jurisdiction, not GPU location. A regional gateway hostname does not pin a downstream GPU. For a binding residency requirement, confirm the model, region, failover scope and enforced controls before sending sensitive data.
Measured performance
34 samplesContinuously sampled across IO Intelligence'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 | 1493 ms |
|---|---|
| Effective throughput | 65 tok/s n=12 |
| Uptime | 100.00% |
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| mistralai/mistral-nemo-instruct-2407 | 881 ms | — | 100.00% | — | 2 |
| moonshotai/kimi-k2.5 | 977 ms | — | 100.00% | — | 1 |
| openai/gpt-oss-20b | 1084 ms | — | 100.00% | — | 2 |
| qwen/qwen3.8-27b | 1093 ms | — | 100.00% | — | 2 |
| minimax/minimax-m2.5 | 1155 ms | — | 100.00% | — | 2 |
| meta-llama/llama-4-maverick-17b-128e-instruct-fp8 | 1218 ms | — | 100.00% | — | 2 |
| deepseek/deepseek-v4-flash | 1278 ms | 67 tok/s n=1 | 100.00% | — | 1 |
| deepseek/deepseek-r1-0528 | 1289 ms | — | 100.00% | — | 1 |
| moonshotai/kimi-k3 | 1400 ms | 50 tok/s n=2 | 100.00% | — | 1 |
| z-ai/glm-5 | 1465 ms | — | 100.00% | — | 1 |
| google/gemma-4-26b-a4b-it | 1493 ms | — | 100.00% | — | 1 |
| moonshotai/kimi-k2-thinking | 1570 ms | — | 100.00% | — | 2 |
| z-ai/glm-4.7 | 1671 ms | — | 100.00% | — | 2 |
| moonshotai/kimi-k2.6 | 1769 ms | 41 tok/s n=2 | 100.00% | — | 4 |
| qwen/qwen3-next-80b-a3b-instruct | 1911 ms | — | 100.00% | — | 1 |
| z-ai/glm-4.5-air | 2024 ms | — | 100.00% | — | 1 |
| google/gemma-4-31b-it | 2352 ms | 114 tok/s n=2 | 100.00% | — | 1 |
| z-ai/glm-5.1 | 3325 ms | — | 100.00% | — | 1 |
| moonshotai/kimi-k2.7-code | 3439 ms | 137 tok/s n=2 | 100.00% | — | 1 |
| z-ai/glm-5.3 | 3504 ms | 62 tok/s n=2 | 100.00% | — | 1 |
| xiaomi/mimo-v2.6-flash | 5476 ms | — | 100.00% | — | 1 |
| z-ai/glm-4.7-flash | — | — | 100.00% | — | 3 |
| z-ai/glm-5.2 | — | 229 tok/s n=1 | — | — | 0 |
IO Intelligence performance history · Full provider & model leaderboard.
Models served by IO Intelligence.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Input | Cached input | Output |
|---|---|---|---|---|---|
deepseek/deepseek-r1-0528DeepSeek: R1 0528 |
— | 163,840 | $0.598977/1M | $0.299489/1M | $2.404345/1M |
deepseek/deepseek-v3.2DeepSeek: DeepSeek V3.2 |
IQ 102#105 | 163,840 | $1.474637/1M | $0.737319/1M | $2.369446/1M |
deepseek/deepseek-v4-flashDeepSeek: DeepSeek V4 Flash 0423 |
IQ 122#35 | 1,048,576 | $0.181946/1M | $0.090973/1M | $0.709382/1M |
deepseek/deepseek-v4-flash-0731DeepSeek: DeepSeek V4 Flash 0731 |
— | 1,048,576 | $0.237586/1M | $0.118793/1M | $0.628358/1M |
deepseek/deepseek-v4-proDeepSeek: DeepSeek V4 Pro 0423 |
IQ 124#29 | 1,048,576 | $1.47911/1M | $0.739555/1M | $4.78548/1M |
deepseek/deepseek-v4.1-flashDeepSeek: DeepSeek V4.1 Flash |
IQ 126#21 | 1,048,576 | $0.3587/1M | $0.17935/1M | $1.40315/1M |
google/gemma-4-26b-a4b-itGoogle: Gemma 4 26B A4B |
IQ 100#113 | 262,144 | $0.10339/1M | $0.051695/1M | $0.33971/1M |
google/gemma-4-31b-itGoogle: Gemma 4 31B |
IQ 101#110 | 262,144 | $0.4009/1M | $0.20045/1M | $1.21325/1M |
intel/qwen3-coder-480b-a35b-instruct-int4-mixed-arIntel: Qwen3 Coder 480B A35B Instruct INT4 Mixed AR |
— | 106,000 | $0.486619/1M | $0.24331/1M | $2.419907/1M |
meta-llama/llama-3.3-70b-instructMeta: Llama 3.3 70B Instruct |
— | 131,072 | $0.576663/1M | $0.288332/1M | $1.078843/1M |
meta-llama/llama-4-maverick-17b-128e-instruct-fp8Llama 4 Maverick Instruct |
— | 1,048,576 | $0.283972/1M | $0.141987/1M | $0.880047/1M |
minimax/minimax-m2.5MiniMax: MiniMax M2.5 |
IQ 104#98 | 196,608 | $0.36714/1M | $0.18357/1M | $1.44113/1M |
mistralai/mistral-nemo-instruct-2407mistralai/Mistral-Nemo-Instruct-2407 |
— | 131,072 | $0.054227/1M | $0.027114/1M | $0.08651/1M |
moonshotai/kimi-k2-instruct-0905moonshotai/Kimi-K2-Instruct-0905 |
— | 262,144 | $0.60135/1M | $0.300675/1M | $2.4265/1M |
moonshotai/kimi-k2-thinkingMoonshotAI: Kimi K2 Thinking |
IQ 92#146 | 262,144 | $0.633/1M | $0.3165/1M | $2.6375/1M |
moonshotai/kimi-k2.5MoonshotAI: Kimi K2.5 |
IQ 110#74 | 262,144 | $0.568012/1M | $0.284006/1M | $2.990925/1M |
moonshotai/kimi-k2.6MoonshotAI: Kimi K2.6 |
IQ 115#52 | 262,144 | $0.778801/1M | $0.389401/1M | $3.66296/1M |
moonshotai/kimi-k2.7-codeMoonshotAI: Kimi K2.7 Code |
IQ 112#66 | 262,144 | $1.074066/1M | $0.537033/1M | $4.77915/1M |
moonshotai/kimi-k3MoonshotAI: Kimi K3 |
IQ 127#20 | 1,048,576 | $3.79589/1M | $1.897945/1M | $18.99/1M |
nvidia/nemotron-3.5-lightningNVIDIA: Nemotron 3.5 Lightning |
— | 262,144 | $0.067257/1M | $0.033629/1M | $0.160888/1M |
openai/gpt-oss-120bOpenAI: gpt-oss-120b |
IQ 98#126 | 131,072 | $0.18779/1M | $0.093895/1M | $0.7174/1M |
openai/gpt-oss-20bOpenAI: gpt-oss-20b |
IQ 94#140 | 131,072 | $0.058025/1M | $0.029013/1M | $0.20467/1M |
qwen/qwen3-next-80b-a3b-instructQwen: Qwen3 Next 80B A3B Instruct |
— | 262,144 | $0.119567/1M | $0.059784/1M | $1.195667/1M |
qwen/qwen3.6-27bQwen: Qwen3.6 27B |
IQ 111#72 | 262,144 | $0.355535/1M | $0.177768/1M | $3.13335/1M |
qwen/qwen3.6-35b-a3bQwen: Qwen3.6 35B A3B |
IQ 104#100 | 262,144 | $0.184836/1M | $0.092418/1M | $1.279452/1M |
qwen/qwen3.8-27bQwen: Qwen3.8 27B |
IQ 116#50 | 65,536 | $0.48319/1M | $0.241595/1M | $2.967715/1M |
xiaomi/mimo-v2.5Xiaomi: MiMo-V2.5 |
IQ 108#82 | 1,050,000 | $0.209945/1M | $0.104973/1M | $0.67309/1M |
xiaomi/mimo-v2.6-flashXiaomi: MiMo-V2.6-Flash |
IQ 122#38 | 1,050,000 | $0.152975/1M | $0.076488/1M | $0.31017/1M |
z-ai/glm-4.5-airZ.ai: GLM 4.5 Air |
— | 131,072 | $0.165284/1M | $0.082643/1M | $0.988184/1M |
z-ai/glm-4.6Z.ai: GLM 4.6 |
— | 202,752 | $0.5486/1M | $0.2743/1M | $2.149563/1M |
z-ai/glm-4.7Z.ai: GLM 4.7 |
IQ 105#90 | 202,752 | $0.59924/1M | $0.29962/1M | $2.24293/1M |
z-ai/glm-4.7-flashZ.ai: GLM 4.7 Flash |
— | 131,072 | $0.066993/1M | $0.033497/1M | $0.422/1M |
z-ai/glm-5Z.ai: GLM 5 |
IQ 107#83 | 202,752 | $0.9073/1M | $0.45365/1M | $2.90336/1M |
z-ai/glm-5.1Z.ai: GLM 5.1 |
IQ 113#59 | 202,752 | $1.477296/1M | $0.738648/1M | $4.642929/1M |
z-ai/glm-5.2Z.ai: GLM 5.2 |
IQ 119#45 | 1,048,576 | $1.86946/1M | $0.93473/1M | $6.12322/1M |
z-ai/glm-5.3Z.ai: GLM 5.3 |
IQ 125#24 | 1,048,576 | $1.7513/1M | $0.87565/1M | $5.5704/1M |
z-ai/glm-5.3-flashZ.ai: GLM 5.3 Flash |
IQ 120#42 | 1,048,576 | $0.152975/1M | $0.076488/1M | $0.80602/1M |
Questions
Does IO Intelligence have zero data retention?
TrustedRouter does not currently mark IO Intelligence as provider-level zero data retention. Use trustedrouter/zdr or provider.min_privacy=zdr to select a different eligible route, and review the linked policy source for changes.
Is IO Intelligence end-to-end encrypted?
TrustedRouter does not currently mark IO Intelligence 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 IO Intelligence models are available through TrustedRouter?
This page currently lists 37 public IO Intelligence models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.