Makora
Try Makora on us. Starter credit covers your first requests, then one key reaches all 7 Makora models, with the privacy posture shown on every route.
Makoramakora
These privacy labels describe Makora, 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 | Makora |
|---|---|
| API operator country | US Legal home of the API operator, not an inference-location guarantee. |
| Headquarters | Not verified |
| Routing status | Active |
| Provider website | https://www.makora.com/ |
| Models | 7 public models |
| Credits routes | 7 |
| Zero data retention | not claimed |
| Verified confidential inference | Not verified |
| Policy note | Makora's Developer Pro and pay-as-you-go tiers advertise ZDR, but permit encrypted on-host SSD caches with no fixed deletion deadline and retention of flagged content for safety investigations. Starter and Developer tiers retain content for 30 days. No training on any tier. These routes remain excluded from strict ZDR. 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
- United States (primary processing region)
Privacy policy section 7 lists AWS and TensorWave inference compute in the US; section 10 also permits international processing. - Can locations change?
- The policy identifies a primary region, not an exhaustive country set or an exact-model guarantee.
- Provider region pinning
- Not verified.
- Pinning through TrustedRouter
- Not supported by the current TrustedRouter integration.
- Infrastructure declaration
- United States primary processing, provider-declared; international processing is permitted and exact-model residency is unverified.
- Evidence
- Public privacy policy sections 7 and 10. 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
27 samplesContinuously sampled across Makora'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 | 1001 ms |
|---|---|
| Effective throughput | 265 tok/s n=3 |
| Uptime | 100.00% |
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| z-ai/glm-5.3 | 963 ms | 265 tok/s n=2 | 100.00% | — | 2 |
| meta-models/muse-spark-open-816b-a42b-nvfp4 | 980 ms | — | 100.00% | 4 provider_error |
5 |
| deepseek/deepseek-v4.1-flash | 1001 ms | — | 100.00% | 2 client_error |
7 |
| xiaomi/mimo-v2.6-flash-rl | 1193 ms | — | 100.00% | — | 2 |
| z-ai/glm-5.3-flash | 1273 ms | — | 100.00% | — | 3 |
| deepseek/deepseek-v4.1-flash-1017 | 1367 ms | — | 100.00% | — | 3 |
| moonshotai/kimi-k3 | 1999 ms | 29 tok/s n=1 | 100.00% | — | 5 |
Makora performance history · Full provider & model leaderboard.
Models served by Makora.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Input | Cached input | Output |
|---|---|---|---|---|---|
deepseek/deepseek-v4.1-flashDeepSeek: DeepSeek V4.1 Flash |
IQ 126#24 | 1,048,576 | $0.2743/1M | $0.01/1M | $1.23435/1M |
deepseek/deepseek-v4.1-flash-1017DeepSeek: DeepSeek V4.1 Flash 1017 |
IQ 128#19 | 1,048,576 | $0.15825/1M | $0.01/1M | $1.266/1M |
meta-models/muse-spark-open-816b-a42b-nvfp4meta-models/Muse-Spark-Open-816B-A42B-NVFP4 |
— | 1,048,576 | $1.31875/1M | $0.15825/1M | $4.48375/1M |
moonshotai/kimi-k3MoonshotAI: Kimi K3 |
IQ 127#22 | 1,048,576 | $0.68575/1M | $0.24265/1M | $15.825/1M |
xiaomi/mimo-v2.6-flash-rlXiaomiMiMo/MiMo-V2.6-Flash-RL |
— | 1,048,576 | $0.13715/1M | $0.01/1M | $0.2954/1M |
z-ai/glm-5.3Z.ai: GLM 5.3 |
IQ 124#30 | 1,048,576 | $0.1477/1M | $0.1055/1M | $4.642/1M |
z-ai/glm-5.3-flashZ.ai: GLM 5.3 Flash |
IQ 119#50 | 1,048,576 | $0.1477/1M | $0.02532/1M | $0.49585/1M |
Questions
Does Makora have zero data retention?
TrustedRouter does not currently mark Makora 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 Makora end-to-end encrypted?
TrustedRouter does not currently mark Makora 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 Makora models are available through TrustedRouter?
This page currently lists 7 public Makora models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.