Skip to content

[ Pricing ]

Pay per token, nothing else.

Per-token, pay as you go. No subscriptions and no minimums; you pay only for the tokens you use.

USD per 1M tokens

Model prices
Modalities
101790% offtext and image in, text outDeepSeek (official): $0.30; Engy: $0.03 per million tokensDeepSeek (official): $1.20; Engy: $0.12 per million tokensDeepSeek (official): $0.006; Engy: $0.006 per million tokens90% off1,048,576 tokens
—text in, text outEngy: $0.045 per million tokensEngy: $0.09 per million tokensEngy: $0.009 per million tokens—1,048,576 tokens
87% offtext in, text outQwen (official): $0.375; Engy: $0.045 per million tokensQwen (official): $2.25; Engy: $0.30 per million tokensEngy: $0.015 per million tokens87% off208,192 tokens
89% offtext and image in, text outQwen (official): $0.50; Engy: $0.045 per million tokensQwen (official): $3.00; Engy: $0.32 per million tokensEngy: $0.015 per million tokens89% off1,001,536 tokens
10% offtext and image in, text outZ.ai (official): $0.15; Engy: $0.135 per million tokensZ.ai (official): $0.50; Engy: $0.45 per million tokensZ.ai (official): $0.03; Engy: $0.027 per million tokens10% off327,680 tokens
58% offtext in, text outZ.ai (official): $1.40; Engy: $0.68 per million tokensZ.ai (official): $4.40; Engy: $1.50 per million tokensZ.ai (official): $0.26; Engy: $0.18 per million tokens58% off262,144 tokens
30% offtext in, text outZ.ai (official): $1.40; Engy: $0.98 per million tokensZ.ai (official): $4.40; Engy: $3.08 per million tokensZ.ai (official): $0.26; Engy: $0.18 per million tokens30% off327,680 tokens
35% offtext and image in, text outMoonshot (official): $3.00; Engy: $1.95 per million tokensMoonshot (official): $15.00; Engy: $9.75 per million tokensMoonshot (official): $0.30; Engy: $0.195 per million tokens35% off1,113,088 tokens
—unknown: not reported in engy's catalogEngy: not applicableEngy: $0.03 per secondEngy: not applicable—unknown: not reported in engy's catalog

Discount shows the percentage saved compared with the developer's official API, using a 3:1 input-to-output token mix. It updates with the catalog price; “more” means engy costs more. A dash means no comparable official price is available.

Prompt-cache hits bill at the cached rate automatically, with no config and no cache_control markers. Agentic workloads such as coding assistants and multi-turn tools reuse long prefixes, so the effective input cost is usually well below the headline rate.

The engy gateway does not store your prompts or outputs, on any model. For zero data retention end to end, set the ZDR flag on a request. See terms and privacy.

These are the catalog's list prices. The API reports the same numbers at https://api.engy.ai/v1/models.