Kimi K3.1 Pro, GLM-5.3, DeepSeek V4 and Qwen 3.8 on an OpenAI-compatible endpoint. Run on hardware we own and operate, with the price published in full.
1M context · 99.9% monthly availability target · zero retention by default · rate 1.00
# OpenAI-compatible. Change the base URL. from openai import OpenAI client = OpenAI( base_url="https://api.tokentoken.ai/v1", api_key="tt-...", ) r = client.chat.completions.create( model="tt/kimi-k3.1-pro", messages=[{"role": "user", "content": "hi"}], )
Anthropic Messages supported on the same key.
Frontier open weights on our own hardware. Vendor APIs are being onboarded.
| Model | Weights | Context | Input /M | Output /M | Status |
|---|---|---|---|---|---|
| tt/kimi-k3.1-pro | Kimi K3.1 Pro | 1M | $2.00 | $10.00 | Available |
| tt/glm-5.3 | GLM-5.3 | 1M | $1.40 | $4.40 | Available |
| tt/deepseek-v4-pro | DeepSeek V4-Pro | 1M | $1.32 | $3.96 | Available |
| tt/deepseek-v4.1-flash | DeepSeek V4.1-Flash | 1M | $0.30 | $1.20 | Available |
| tt/glm-5.3-flash | GLM-5.3-Flash | 1M | $0.15 | $0.50 | Available |
| tt/qwen3.8-27b | Qwen3.8-27B | 262K | $0.50 | $3.00 | Available |
| claude-opus-5-5 | Anthropic API | 1M | — | — | Coming soon |
| gpt-6-astra | OpenAI API | — | — | — | Coming soon |
| gemini-4-argon | Google API | — | — | — | Coming soon |
Prices at rate 1.00. For each set of open weights we charge what the weights author charges on their own API — no markup, and no undercutting we cannot yet justify. Vendor-API models will list at the vendor's published rate while we onboard them.
Frontier open weights sit an order of magnitude below frontier closed models. Inside that band we charge the weights author's own published rate — rate 1.00, no markup.
| Model | Source | Input /M | Output /M |
|---|---|---|---|
| tt/deepseek-v4.1-flash | Self-hosted — DeepSeek's published rate | $0.30 | $1.20 |
| tt/deepseek-v4-pro | Self-hosted — DeepSeek's published rate | $1.32 | $3.96 |
| tt/glm-5.3 | Self-hosted — Z.ai's published rate | $1.40 | $4.40 |
| tt/kimi-k3.1-pro | Self-hosted — Moonshot's published rate | $2.00 | $10.00 |
| gemini-4-argon | Google published rate | $2.00 | $10.00 |
| claude-opus-5-5 | Anthropic published rate | $4.00 | $20.00 |
| gpt-6-astra | OpenAI published rate | $10.00 | $50.00 |
Closed-model rates are each vendor's published standard-tier price inside their short-context band. Our self-hosted prices are the published rate of the same weights' author, read 6–7 November 2026: Moonshot for Kimi K3.1 Pro, Z.ai for GLM-5.3, DeepSeek's peak-hour rate for V4-Pro and V4.1-Flash. DeepSeek also publishes an off-peak rate at half of that; we do not price by time of day.
One team owns the kernels, the scheduler, the fleet and the pager. That is what makes a route stay up under load. We charge the weights author's own published rate rather than a markup on someone else's quota — so what you are buying from us is the capacity, the reliability and the data handling, not a discount.
One key, one endpoint. The gateway authorizes, reserves budget and routes; the fleet produces the tokens.
Self-hosted routes are live. Vendor-API routes are being onboarded and are marked coming soon in the catalogue and in GET /v1/models.
Our own GPUs, not resold quota. Reserved hardware is yours alone, and its idle time feeds the token pool.
Kernels tuned per card, quantization, speculative decoding, continuous batching. Throughput per GPU is our own engineering.
Every failure path is detect → remediate → verify, each held by a gate that injects the fault and asserts the fix fires.
Every price is a base price times rate, now 1.00 — on this site, in the console and in the API. The number we bill.
Most teams start on the API and pay per token. When your load is steady enough to keep a card busy, reserve one: B300 in stock, by the month, power and bandwidth included. We quote within one business day.
What fits without splitting the model across cards.
HBM3e on B300 and H200; GDDR7 and GDDR6X on the RTX cards. Bars are to scale.
Prompts and outputs are not stored. Billing metadata is. Debug capture is opt-in per key and expires on its own.
No setting changes this and no discount is offered for allowing it.
Self-hosted, vendor API, cloud or licensed third party — in the API response, not only on this page. No account pools.
Selected routes run in a hardware TEE and return an attestation your client verifies before the prompt leaves, plus a signed usage receipt.
Inference can be contracted to stay in North America, Southeast Asia or East Asia.
Authorization headers are dropped at the entry point, not redacted afterwards.
DPA with SCCs and UK Addendum, sub-processor list with 30 days' notice, security overview.