Your agent pays for its own LLM calls in stablecoins. OpenAI-compatible API, per-request x402 payments in USDC on Base, or prepaid tai_ keys topped up on-chain. No card, no account, no signup.
from openai import OpenAI
client = OpenAI(base_url="https://pay.trustai.center/v1", api_key="tai_...")
r = client.chat.completions.create(model="deepseek-v4.1-flash",
messages=[{"role": "user", "content": "Hello"}])Drop-in /v1/chat/completions and /v1/models. Streaming supported. Up to 16k output tokens (or the model max), ~400k chars of input, 500 messages. Per-key spend cap, daily limit and per-request max.
Call without a key and get an HTTP 402 with payment requirements. Pay USDC on Base per request; settlement happens only after a successful answer.
https://pay.trustai.center/mcp — list_models, list_trending, get_balance, get_usage, set_cap, create_topup, verify_topup, chat_completion, get_stats.
USDC on Base or Solana, USDT/USDC on BNB Chain. The agent asks for an address and exact amount, pays, verifies. Credits land on its key.
Live prices, updated 2026-10-25 20:15 UTC.
| Model | Input $/1M | Output $/1M | List (in / out) | vs list | Max out |
|---|---|---|---|---|---|
deepseek-v4.1-flash #1 | $0.15 | $0.6 | $0.15 / $0.6 | at list price ✓ | 16.0k |
glm-5.3-flash #2 | $0.15 | $0.5 | $0.15 / $0.5 | at list price ✓ | 16.0k |
qwen3-235b #3 | $0.9 | $0.9 | $0.9 / $0.9 | at list price ✓ | 16.0k |
gpt-oss-120b #4 | $0.1 | $0.5 | $0.1 / $0.5 | at list price ✓ | 16.0k |
deepseek-v4.1-flash-1017 #5 | $0.15 | $0.6 | $0.15 / $0.6 | at list price ✓ | 16.0k |
qwen3.7-flash #6 | $0.03 | $0.13 | $0.03 / $0.13 | at list price ✓ | 16.0k |
glm-4.7-flash #7 | $0.07 | $0.4 | $0.07 / $0.4 | at list price ✓ | 16.0k |
kimi-k2-thinking #8 | $0.47 | $2 | $0.47 / $2 | at list price ✓ | 16.0k |
llama-4-scout #9 | $0.1 | $0.3 | $0.1 / $0.3 | at list price ✓ | 8.2k |
claude-haiku-5.5 #10 | $0.1 | $0.5 | $0.1 / $0.5 | at list price ✓ | 16.0k |
llama-3.1-8b | $0.02 | $0.04 | $0.02 / $0.04 | at list price ✓ | 8.2k |
gemma-3-4b-it | $0.05 | $0.1 | $0.05 / $0.1 | at list price ✓ | 8.0k |
ministral-3b | $0.1 | $0.1 | $0.1 / $0.1 | at list price ✓ | 4.0k |
ling-3.0-flash-fin | $0.042 | $0.123 | $0.042 / $0.123 | at list price ✓ | 16.0k |
ling-3.0-flash-sante | $0.042 | $0.123 | $0.042 / $0.123 | at list price ✓ | 16.0k |
e2ee-qwen-2.5-7b-p | $0.05 | $0.13 | $0.05 / $0.13 | at list price ✓ | 16.0k |
mistral-nemo #13 | $0.15 | $0.15 | $0.15 / $0.15 | at list price ✓ | 16.0k |
gemma-3-12b-it | $0.05 | $0.15 | $0.05 / $0.15 | at list price ✓ | 8.2k |
mercury-2.5 | $0.04 | $0.15 | $0.04 / $0.15 | at list price ✓ | 16.0k |
ministral-8b | $0.15 | $0.15 | $0.15 / $0.15 | at list price ✓ | 4.0k |
qwen3.5-9b | $0.1 | $0.15 | $0.1 / $0.15 | at list price ✓ | 16.0k |
ling-3.0-flash | $0.06 | $0.18 | $0.06 / $0.18 | at list price ✓ | 16.0k |
ling-3.0-flash-vl | $0.06 | $0.18 | $0.06 / $0.18 | at list price ✓ | 16.0k |
gpt-oss-safeguard-20b | $0.07 | $0.2 | $0.07 / $0.2 | at list price ✓ | 16.0k |
ministral-14b | $0.2 | $0.2 | $0.2 / $0.2 | at list price ✓ | 16.0k |
mistral-small-3.2-24b-instruct | $0.075 | $0.2 | $0.075 / $0.2 | at list price ✓ | 16.0k |
muse-spark-1.2-contributor | $0.1 | $0.2 | $0.1 / $0.2 | at list price ✓ | 16.0k |
muse-spark-1.3-contributor | $0.1 | $0.2 | $0.1 / $0.2 | at list price ✓ | 16.0k |
nemotron-nano-12b-v2-vl | $0.2 | $0.2 | $0.2 / $0.2 | at list price ✓ | 16.0k |
qwen3-14b | $0.08 | $0.2 | $0.08 / $0.2 | at list price ✓ | 16.0k |
247 servable chat models, prices synced daily from the model makers' list prices: we sell at list price (never more than 2% above list, ✓ = verified on this page). Full catalog: /v1/models · discovery (image/video, not served): ?include=discovery.
tokens served (24h)
requests (24h)
tokens served (all time)
requests (all time)
API keys created
Top models by our usage (7d): gpt-oss-120b, claude-sonnet-4.6, deepseek-v4.1-flash, glm-5.3-flash
Real numbers from our database: /v1/stats
Machine-readable: /v1/models · /llms.txt · /.well-known/trustai.json · /.well-known/mcp.json
Promo: code LAUNCH20 gives +20% credits on the first top-up of a key, until 1 Nov 2026 23:59 Paris time.
Top models by tokens served in the last 24h across our inference network, refreshed daily. Full catalog: /v1/models (servable chat models by default; ?include=discovery adds image/video listed for reference only; filters: ?modality=text, &max_price=2, &sort=popularity|price, &q=qwen).
deepseek-v4.1-flash $0.15 in / $0.6 out per 1M · at list priceglm-5.3-flash $0.15 in / $0.5 out per 1M · at list priceqwen3-235b $0.9 in / $0.9 out per 1M · at list pricegpt-oss-120b $0.1 in / $0.5 out per 1M · at list pricedeepseek-v4.1-flash-1017 $0.15 in / $0.6 out per 1M · at list priceqwen3.7-flash $0.03 in / $0.13 out per 1M · at list priceglm-4.7-flash $0.07 in / $0.4 out per 1M · at list pricekimi-k2-thinking $0.47 in / $2 out per 1M · at list pricellama-4-scout $0.1 in / $0.3 out per 1M · at list priceclaude-haiku-5.5 $0.1 in / $0.5 out per 1M · at list price