Pricing
Pay per token. Nothing else.
No subscriptions, no minimums, no seat fees. Every request is billed from your balance at the rates below, per 1,000,000 tokens. Cached prompt tokens are billed at a discount.
Cheapest model
DeepSeek V4.1 Flash · $0.04/1M
Commitment
None — cancel anytime
Billing model
Prepaid balance, metered live
Gemini
1 models · Long-context multimodal work.
| Model | Context | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|---|
| gemini-4-argon | 128K | $0.3 | $0.08 | $0 |
GPT
8 models · General reasoning, tool use and agents.
| Model | Context | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|---|
| gpt-6-astra | 1.1M | $0.9 | $0.08 | $0 |
| gpt-6.1-sol | 1.1M | $0.3 | $0.08 | $0 |
| gpt-6-sol | 1.1M | $0.3 | $0.08 | $0 |
| gpt-6-luna | 1.1M | $0.1 | $0.02 | $0 |
| gpt-5.6-sol | 1M | $0.9 | $0.08 | $0 |
| gpt-5.6-terra | 1M | $0.9 | $0.08 | $0 |
| gpt-5.6-luna | 1M | $0.9 | $0.08 | $0 |
| gpt-5.5 | 1M | $0.9 | $0.08 | $0 |
Claude
9 models · Long-horizon reasoning and careful writing.
| Model | Context | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|---|
| claude-fable-5 | 1M | $0.9 | $0.08 | $0 |
| claude-haiku-5-5 | 1M | $0.06 | $0.02 | $0 |
| claude-sonnet-5 | 1M | $0.9 | $0.08 | $0 |
| claude-sonnet-5-5 | 1M | $1.00 | $0.12 | $0 |
| claude-opus-4-6 | 1M | $0.9 | $0.08 | $0 |
| claude-opus-4-7 | 1M | $0.9 | $0.08 | $0 |
| claude-opus-4-8 | 1M | $0.9 | $0.08 | $0 |
| claude-opus-5 | 1M | $0.9 | $0.08 | $0 |
| claude-opus-5-5 | 1M | $1.40 | $0.14 | $0 |
Grok
3 models · Real-time aware chat and reasoning.
DeepSeek
4 models · Cost-efficient coding, math and open weights.
| Model | Context | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|---|
| deepseek-v4.1-flash | 1M | $0.04 | $0.01 | $0 |
| deepseek-v4-flash-0731 | 1M | $0.06 | $0.0053 | $0 |
| deepseek-v4-pro-0813 | 1M | $0.5 | $0.1 | $0 |
DeepSeek V4 Pro UncensoredUncensored deepseek-v4-pro-uncensored | 1M | $0.5 | $0.1 | $0 |
GLM
7 models · Bilingual work plus uncensored and abliterated tiers.
| Model | Context | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|---|
| glm-5.2 | 1M | $0.12 | $0.03 | $0 |
| glm-5.3 | 1M | $0.12 | $0.03 | $0 |
GLM-5.3 UncensoredUncensored glm-5.3-uncensored | 1M | $0.12 | $0.03 | $0 |
| glm-5.3-flash | 1M | $0.06 | $0.0056 | $0 |
GLM-5.3 Flash UncensoredUncensored glm-5.3-flash-uncensored | 1M | $0.06 | $0.0056 | $0 |
GLM-5.3 AbliteratedAbliterated glm-5.3-abliterated | 1M | $1.50 | $0.44 | $0 |
GLM-5.3 Flash AbliteratedAbliterated glm-5.3-flash-abliterated | 1M | $0.44 | $0.12 | $0 |
MiMo
2 models · Efficient generalist tuned for high throughput.
| Model | Context | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|---|
| mimo-v2.6-pro | 600K | $0.2 | $0.0060 | $0 |
| mimo-v2.6-flash | 600K | $0.06 | $0.0040 | $0 |
Kimi
1 models · Agentic workflows and retrieval-heavy tasks.
| Model | Context | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|---|
| kimi-k3 | 1M | $0.3 | $0.06 | $0 |
Input tokens
Billed per 1M uncached prompt tokens. The full conversation you send counts here.
Cached tokens
Repeated prefixes served from cache are billed at the discounted cached rate.
Output tokens
Completion tokens are billed at the output rate — set per model in the catalog.
Start building today
Generate a key and make your first request in under a minute.