Lyceum
Explore Lyceum models on TrustedRouter with current routes, token pricing, policy sources, privacy posture, regional availability, and API support.
Lyceumlyceum
These privacy labels describe Lyceum, 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 | Lyceum |
|---|---|
| API operator country | DE Legal home of the API operator, not an inference-location guarantee. |
| Headquarters | Not verified |
| Routing status | Active |
| Provider website | https://lyceum.technology/ |
| Models | 13 public models |
| Credits routes | 13 |
| Zero data retention | not claimed |
| Verified confidential inference | Not verified |
| Policy note | Lyceum's terms state that inference inputs and outputs are used only to generate results, are not stored beyond technical necessity, and are not used for training or analysis. No account-specific ZDR or verified confidential-compute guarantee is tracked; these routes remain Standard. German company jurisdiction is not a guarantee that every model executes in the EU. 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.
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
- Not verified
Lyceum documents model-specific serving locations; its German headquarters is not a fleet-wide EU residency guarantee. - Can locations change?
- Placement and cross-region failover depend on the model; an exhaustive country list is not verified.
- Provider region pinning
- Not verified.
- Pinning through TrustedRouter
- Not supported by the current TrustedRouter integration.
- Infrastructure declaration
- Inference location not verified; no country-specific residency commitment.
- Evidence
- Public provider model-roster documentation. 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
31 samplesContinuously sampled across Lyceum'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 | 1505 ms |
|---|---|
| Effective throughput | 112 tok/s n=7 |
| Uptime | 96.77% |
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| z-ai/glm-5.3 | 929 ms | 112 tok/s n=1 | 100.00% | — | 2 |
| deepseek/deepseek-v4-flash-0731 | 1157 ms | — | 100.00% | — | 4 |
| z-ai/glm-5.2 | 1343 ms | 122 tok/s n=2 | 100.00% | — | 4 |
| z-ai/glm-5.3-flash | 1438 ms | — | 100.00% | — | 3 |
| deepseek/deepseek-v4-pro | 1505 ms | — | 100.00% | — | 3 |
| moonshotai/kimi-k3 | 1538 ms | 112 tok/s n=2 | 100.00% | — | 2 |
| moonshotai/kimi-k2.7-code | 1680 ms | 154 tok/s n=1 | 100.00% | — | 2 |
| qwen/qwen3.8-flash-next | 1710 ms | — | 100.00% | — | 2 |
| qwen/qwen3.8-27b | 1932 ms | — | 100.00% | — | 2 |
| minimax/minimax-m3 | 2661 ms | 29 tok/s n=1 | 100.00% | — | 3 |
| z-ai/glm-5.1 | 3127 ms | — | 100.00% | — | 1 |
| qwen/qwen3.8-2.4t-a95b | 3216 ms | — | 100.00% | — | 2 |
| moonshotai/kimi-k2.6 | — | — | 0.00% | 1 unsupported_route |
1 |
| qwen/qwen3-235b-a22b-instruct-2507 | — | — | — | 2 unsupported_route |
0 |
Lyceum performance history · Full provider & model leaderboard.
Models served by Lyceum.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Input | Cached input | Output |
|---|---|---|---|---|---|
deepseek/deepseek-v4-flash-0731DeepSeek: DeepSeek V4 Flash 0731 |
— | 1,048,576 | $0.26375/1M | $0.0633/1M | $0.3165/1M |
deepseek/deepseek-v4-proDeepSeek: DeepSeek V4 Pro 0423 |
IQ 123#30 | 1,048,576 | $1.84625/1M | $0.4642/1M | $3.6925/1M |
deepseek/deepseek-v4.1-flashDeepSeek: DeepSeek V4.1 Flash |
IQ 125#22 | 1,048,576 | $0.5275/1M | $0.13715/1M | $1.5825/1M |
minimax/minimax-m3MiniMax: MiniMax M3 |
IQ 117#45 | 524,288 | $0.422/1M | $0.1055/1M | $2.11/1M |
moonshotai/kimi-k2.7-codeMoonshotAI: Kimi K2.7 Code |
IQ 111#66 | 262,144 | $1.31875/1M | $0.32705/1M | $4.7475/1M |
moonshotai/kimi-k3MoonshotAI: Kimi K3 |
IQ 128#17 | 1,048,576 | $3.165/1M | $0.79125/1M | $15.825/1M |
qwen/qwen3-embedding-8bQwen3 Embedding 8B |
— | 32,768 | $0.0211/1M | Not published | selected route |
qwen/qwen3.8-2.4t-a95bQwen: Qwen3.8 2.4T A95B |
IQ 124#29 | 262,144 | $2.6375/1M | $0.66465/1M | $6.33/1M |
qwen/qwen3.8-27bQwen: Qwen3.8 27B |
IQ 116#48 | 65,536 | $0.422/1M | $0.1055/1M | $2.532/1M |
qwen/qwen3.8-flash-nextQwen: Qwen3.8-Flash-Next |
— | 262,144 | $0.211/1M | $0.05275/1M | $0.5275/1M |
z-ai/glm-5.2Z.ai: GLM 5.2 |
IQ 119#42 | 1,048,576 | $1.5825/1M | $0.4009/1M | $4.7475/1M |
z-ai/glm-5.3Z.ai: GLM 5.3 |
IQ 127#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 122#35 | 1,048,576 | $0.211/1M | $0.05275/1M | $0.5275/1M |
Questions
Does Lyceum have zero data retention?
TrustedRouter does not currently mark Lyceum 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 Lyceum end-to-end encrypted?
TrustedRouter does not currently mark Lyceum 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 Lyceum models are available through TrustedRouter?
This page currently lists 13 public Lyceum models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.