CURATED AI MODEL MARKET
Specs, pricing, billing — fully transparent
One API key for text and video models. Specs tested before listing, billed by usage.
Verification specs
On shelf- Spec test
- ✓Matches official spec
- Capability test
- ✓Passed before listing
- Pricing
- $0.75 in · $3.75 out /1M
- Cache price
- $0.075 /1M
- Data boundary
- Billing and audit metadata only
- Spec test
- ✓Matches official spec
- Capability test
- ✓Passed before listing
- Pricing
- $2.50 in · $12.50 out /1M
- Cache price
- $0.0625 /1M
- Data boundary
- Billing and audit metadata only
- Spec test
- ✓Matches official spec
- Render test
- ✓Verified on real billed tasks
- Pricing
- Per task · live estimate before submit
- Flow
- Submit → poll → download
- Data boundary
- Task params and result URL retained
Re-tested before every listing
All 20 models →
QUICK CONNECT
Key to first call in three steps
Get a key, add it to your client, or call the HTTP endpoints directly.
02
Configure your tool
// ~/.claude/settings.json { "env": { "ANTHROPIC_BASE_URL": "https://tryaiapi.com", "ANTHROPIC_AUTH_TOKEN": "sk-your-key" } }
Restart Claude Code after saving
# ~/.codex/config.toml [model_providers.tryaiapi] name = "Try AI API" base_url = "https://tryaiapi.com/v1" env_key = "TRYAIAPI_KEY" [profiles.default] model_provider = "tryaiapi" model = "gpt-6-astra"
Put your key in the TRYAIAPI_KEY environment variable; fields follow the docs
# Cherry Studio Settings → Model services → Add provider Type: OpenAI-compatible API base: https://tryaiapi.com API key: sk-your-key Add model → enter the model ID (e.g. claude-sonnet-5)
Works for most OpenAI-compatible clients
More tools and full parameters in the docs →
03
Make your first call
curl https://tryaiapi.com/v1/messages \ -H "x-api-key: $TRYAIAPI_KEY" \ -H "anthropic-version: 2023-06-01" \ -d '{ "model": "claude-sonnet-5", "max_tokens": 1024, "messages": [{ "role": "user", "content": "..." }] }'
curl https://tryaiapi.com/v1/chat/completions \ -H "Authorization: Bearer $TRYAIAPI_KEY" \ -d '{ "model": "gpt-6-astra", "messages": [{ "role": "user", "content": "..." }] }'
curl https://tryaiapi.com/videos/v1/videos/generations \ -H "Authorization: Bearer $TRYAIAPI_KEY" \ -d '{ "model": "doubao-seedance-2.0", "prompt": "..." }' # async task: submit → poll → download
The live list is the model market and GET /v1/models
ON THE SHELF
20 on the shelf, pure, authentic
Text chat
Pay as you go16 / 20
fable-5-11M context
75% off+
$2.50input /1M
$12.50output /1M
$0.0625cache /1M
Tested
opus-51M context
75% off+
$1.25input /1M
$6.25output /1M
$0.125cache /1M
Tested
sonnet-51M context
75% off+
$0.75input /1M
$3.75output /1M
$0.075cache /1M
Tested
gpt-6-astra1M context
75% off+
$2.50input /1M
$12.50output /1M
$0.25cache /1M
Tested
gpt-5.6-sol1M context
75% off+
$1.00input /1M
$5.00output /1M
$0.10cache /1M
Tested
grok-4.6
75% off
$0.50input /1M
$1.50output /1M
$0.125cache /1M
Tested
kimi-k31M context
10% off
$2.70input /1M
$13.50output /1M
$0.27cache /1M
Tested
Video generation
Per task · async flow4 / 20
doubao-seedance-2.0
—
Per taskduration × resolution
Live estimatebefore submit
—cache
Tested
doubao-seedance-2.0-fast
—
Per taskduration × resolution
Live estimatebefore submit
—cache
Tested
doubao-seedance-2.5
—
Per taskduration × resolution
Live estimatebefore submit
—cache
Tested
MiniMax-H3
—
Per taskduration × resolution
Live estimatebefore submit
—cache
Tested
All 20 models with live pricing →
Prices match actual billing; the authoritative list is GET /v1/models
Listing changelog
gpt-6-astralisted — 1M context; cache and long-context tiers matched to the official price card
fable-5-1listed — 1M context, thinking supported
grok-4.5 / -4.3 / -4.20 / -buildlisted — the full xAI Grok lineup (6 models), specs tested
wan3.0-videolisted — video async task, specs tested
gpt-5.6-sol / -terra / -lunalisted — all three tiers tested against official specs
claude-sonnet-5listed on the Value tier, billing verified token by token
gpt-image-2listed — per-image billing, Standard / Fine fidelity tiers
doubao-seedance-2.0 / -fastvideo duo restored, verified on real billed tasks
seedance seriesnative Ark-format request parameters published
QUERYABLE
Open books, every entry queryable
§1
Public schema snapshot
Our billing and logging schema is published on the transparency page — open it and see what we store.
/transparency/schema.json →
§2
Usage queryable line by line
Every call is queryable by model, key, and time — built for internal cost allocation and reconciliation.
§3
Minimal metadata logging
We store only what billing, routing, and troubleshooting need.
FAQ
Before you start, you might ask
How do I get started
Sign up in the console, create an API key, and configure your tool per Quick Connect above. Pay as you go — no monthly fee, no minimum spend.
Which protocols are supported
Three endpoints: Anthropic-native /v1/messages, OpenAI-compatible /v1/chat/completions, and async video /videos/v1. The live list is GET /v1/models; full parameters in the docs.
Teams and businesses
We provision team, project, and per-seat quotas manually, with reconciliation and invoicing workflows for small and medium B2B. Email us or reach out in the console.
How is pricing set
Discounts are stated against official list price, itemized across input, output, and cache. Market-page prices match actual billing; the console at order time is authoritative.
Do you store my request content
Successful text requests store no prompts or completions — only billing metadata such as token counts, timestamps, and model IDs. Video jobs temporarily store parameters and download links for polling. Full boundary and schema snapshot on the transparency page.
Who do I reach when something breaks
Email [email protected]. Tag security and privacy issues with [SECURITY] in the subject.
