Add DigitalOcean Gradient AI Platform as a Provider

[Sunny-DotNet]

Request: Add DigitalOcean Gradient AI Platform as a Provider

Overview

DigitalOcean Gradient™ AI Platform offers serverless inference via an OpenAI-compatible API with many foundation models from Anthropic, OpenAI, Arcee, Meta, Mistral, NVIDIA, and more.

• API Endpoint: https://inference.do-ai.run/v1
• Auth: Bearer token (DigitalOcean model access key)
• SDK: @ai-sdk/openai-compatible
• Docs: https://docs.digitalocean.com/products/gradient-ai-platform/

Provider Configuration

name = "DigitalOcean"
npm = "@ai-sdk/openai-compatible"
api = "https://inference.do-ai.run/v1"
env = ["DIGITALOCEAN_API_KEY"]
doc = "https://docs.digitalocean.com/products/gradient-ai-platform/"

Available Models

Source: Pricing | Models

Anthropic Models (8)

Model	Model ID	Input $/1M	Output $/1M	Context	Max Output
Claude Sonnet 4.6	anthropic-claude-4.6-sonnet	3.00	15.00	1M	64K
Claude Sonnet 4.5	anthropic-claude-4.5-sonnet	3.00	15.00	1M	64K
Claude Sonnet 4	anthropic-claude-sonnet-4	3.00	15.00	1M	64K
Claude Haiku 4.5	anthropic-claude-haiku-4.5	1.00	5.00	200K	64K
Claude Opus 4.6	anthropic-claude-opus-4.6	5.00	25.00	1M	128K
Claude Opus 4.5	anthropic-claude-opus-4.5	5.00	25.00	200K	64K
Claude Opus 4.1	anthropic-claude-4.1-opus	15.00	75.00	200K	32K
Claude Opus 4	anthropic-claude-opus-4	15.00	75.00	200K	32K

Arcee Models (1)

Model	Model ID	Input $/1M	Output $/1M	Max Output
Trinity Large	trinity-large-thinking	0.25	0.90	128K

OpenAI Models (20)

Model	Model ID	Input $/1M	Output $/1M	Max Output
GPT-5.4	openai-gpt-5.4	2.50	15.00	128K
GPT-5.4 mini	openai-gpt-5.4-mini	0.75	4.50	128K
GPT-5.4 nano	openai-gpt-5.4-nano	0.20	1.25	128K
GPT-5.4 pro	openai-gpt-5.4-pro	30.00	180.00	128K
GPT-5.3-Codex	openai-gpt-5.3-codex	1.75	14.00	128K
GPT-5.2	openai-gpt-5.2	1.75	14.00	128K
GPT-5.2 pro	openai-gpt-5-2-pro	21.00	168.00	128K
GPT-5.1-Codex-Max	openai-gpt-5.1-codex-max	1.25	10.00	128K
GPT-5	openai-gpt-5	1.25	10.00	128K
GPT-5 mini	openai-gpt-5-mini	0.25	2.00	128K
GPT-5 nano	openai-gpt-5-nano	0.05	0.40	128K
GPT-4.1	openai-gpt-4.1	2.00	8.00	32K
GPT-4o	openai-gpt-4o	2.50	10.00	16K
GPT-4o mini	openai-gpt-4o-mini	0.15	0.60	16K
o1	openai-o1	15.00	60.00	N/A
o3	openai-o3	2.00	8.00	N/A
o3-mini	openai-o3-mini	1.10	4.40	N/A
GPT Image 1	openai-gpt-image-1	5.00	40.00	N/A
GPT Image 1.5	openai-gpt-image-1.5	5.00	10.00	N/A
gpt-oss-120b	openai-gpt-oss-120b	0.10	0.70	131K
gpt-oss-20b	openai-gpt-oss-20b	0.05	0.45	131K

DigitalOcean-Hosted Open Source Models (9)

Model	Model ID	Input $/1M	Output $/1M	Max Output
Qwen3-32B	alibaba-qwen3-32b	0.25	0.55	40K
DeepSeek R1 Distill Llama 70B	deepseek-r1-distill-llama-70b	0.99	0.99	32K
MiniMax M2.5	minimax-m2.5	0.30	1.20	128K
Kimi K2.5	kimi-k2.5	0.50	2.70	32K
Llama 3.3 70B Instruct	llama3.3-70b-instruct	0.65	0.65	128K
Llama 3.1 8B Instruct	llama3-8b-instruct	0.20	0.20	128K
Mistral NeMo	mistral-nemo-instruct-2407	0.30	0.30	128K
Nemotron-3-Super-120B	nvidia-nemotron-3-super-120b	0.30	0.65	N/A
GLM 5	glm-5	1.00	3.20	128K

Notes

• fal models (Fast SDXL, Flux Schnell, Stable Audio, TTS) are excluded — they use non-standard pricing (per compute second / megapixel / character) that doesn't fit the per-token cost schema.
• Many Anthropic models support prompt caching and >200K context pricing tiers.
• Data sourced from official DigitalOcean documentation last updated 2026-04-01.

I'm working on a PR to add this provider with all model definitions.