Hixtube
Try Hixtube on us. Starter credit covers your first requests, then one key reaches the Hixtube model, with its privacy posture shown on the route.
Hixtubehixtube
These privacy labels describe Hixtube, 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 | Hixtube |
|---|---|
| API operator country | Not verified Legal home of the API operator, not an inference-location guarantee. |
| Headquarters | Not verified |
| Routing status | Active |
| Provider website | https://hixtube.com/ |
| Models | 1 public model |
| Credits routes | 1 |
| Zero data retention | yes All routes |
| Verified confidential inference | Not verified |
| Policy note | Hixtube commits to no persistent storage of prompts, outputs or tool content and no training on customer content. Content-free operational and billing metadata remain. Prefix/KV caches can remain in volatile memory until eviction or shutdown; ZDR does not mean immediate per-request memory erasure. The current inference node is hosted by Verda in Finland through Spheron. Standard TLS, not verified confidential computing or provider-side E2EE; the infrastructure operator can access workload memory. Checkpoint identity, context limits and retirement are declared per route; high availability has not been established. Policy source |
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.
usage.provider_usage.gateway_region identifies the TrustedRouter gateway, not the provider's inference location. A value such as europe-west4 does not mean the model ran in Europe. usage.inference_location.serving_region remains null when the provider has not reported it. Older gateways may return the gateway-only field as provider_usage.region.
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
- Finland (EU)
Current DeepSeek inference runs on a single Verda Finland GPU node rented through Spheron. - Can locations change?
- No provider-side geographic failover is configured. Future capacity changes require a new location review.
- Provider region pinning
- The current endpoint has one serving location; no per-request region selector.
- Pinning through TrustedRouter
- Select provider.only=[hixtube] with allow_fallbacks=false; this does not pin TrustedRouter gateway geography.
- Infrastructure declaration
- Finland inference, operator-declared and checked against the hosting allocation; not hardware-attested location.
- Evidence
- Operator hosting allocation and live catalog reviewed October 27, 2026; inference location only.. 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
56 samplesContinuously sampled across Hixtube'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 | 1276 ms |
|---|---|
| Effective throughput | 102 tok/s n=2 |
| Uptime | 100.00% |
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| dealignai/deepseek-v4.1-flash-1017-uncensored-fp8 | 1276 ms | 102 tok/s n=2 | 100.00% | 40 provider_error |
56 |
Hixtube performance history · Full provider & model leaderboard.
Models served by Hixtube.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Input | Cached input | Output |
|---|---|---|---|---|---|
dealignai/deepseek-v4.1-flash-1017-uncensored-fp8DeepSeek V4.1 Flash 1017 UNCENSORED FP8 (selected) |
— | 1,048,576 | $0.68575/1M | $0.68575/1M | $1.52975/1M |
Questions
Does Hixtube have zero data retention?
TrustedRouter records Hixtube 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 Hixtube end-to-end encrypted?
TrustedRouter does not currently mark Hixtube 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 Hixtube models are available through TrustedRouter?
This page currently lists 1 public Hixtube model, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.