Coral Bricks
Fast inference for coding and research agents: DeepSeek V4.1 Flash in three speed tiers (Value, Standard and Fast, which targets over 300 output tok/s) behind an OpenAI-compatible API with Chat Completions, Responses (server-side state) and background mode for long-running agents. 1M-token context on every tier; cached reads are billed from $0.003/1M, with no cache-write charge. Prepaid per-token billing. Custom endpoints on NVIDIA B200/B300 for GLM 5.3, Kimi K3, Gemma 4, GPT-OSS 120B, Qwen 3.6 and more.
At a glance: Coral Bricks has 3 models listed on Inference Hub. The cheapest text model is DeepSeek V4.1 Flash at $0.0050 input and $0.570 output per 1M tokens. There is no free tier. Prices are checked against Coral Bricks's own pricing every week.
Features
Compliance
None listed
Pricing Model
per tokenDetails
https://inference.coralbricks.ai/v1 Model Catalog (5)
| Model | Type | Input $/1M | Output $/1M | Context | Speed | Status |
|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash DeepSeek Fast | llm | $0.450 | $1.80 | 1M | — | |
| DeepSeek V4.1 Flash DeepSeek Standard | llm | $0.190 | $0.600 | 1M | — | |
| DeepSeek V4.1 Flash DeepSeek Value | llm | $0.0050 | $0.570 | 1M | — | |
| GLM 5.3 Z.ai | llm | $1.12 | $4.40 | 1M | — | |
| GLM 5.3 Flash Z.ai | llm | $0.150 | $0.500 | 1M | — |
Coral Bricks API FAQ
Is the Coral Bricks API free?
Not as far as we track: Coral Bricks has no free API tier on record. The cheapest option we track is DeepSeek V4.1 Flash at $0.005 in / $0.57 out per 1M tokens.
What is the cheapest Coral Bricks model?
DeepSeek V4.1 Flash at $0.005 in / $0.57 out per 1M tokens, based on the 5 models we track for Coral Bricks.
How many models does Coral Bricks offer?
We track 5 Coral Bricks models: 5 LLM.
Is the Coral Bricks API OpenAI-compatible?
Yes. Coral Bricks exposes an OpenAI-compatible endpoint, so you can use the OpenAI SDKs by changing the base URL and API key.