Does the Aggregator Support the Exact Same Model List as OpenRouter? Comparing Six Vendors Side by Side
Learn how to read our aggregator's model list for Claude, GPT, DeepSeek, Qwen, GLM, and Kimi, and compare it with OpenRouter's exposed models. Pick the right model ID every time by checking vendor, version, and status before calling.
Why Model IDs Matter
When you call an LLM through an API, you must specify a model ID exactly as the provider expects. A wrong ID returns an error or, worse, silently routes to a different model. Aggregators like ours and OpenRouter each maintain their own model lists, which may differ in naming, versioning, and availability.
How to Read Our Model List
Our model list is the authoritative source for what you can call. Access it via our API or dashboard.
- Vendor and family: Each entry belongs to a vendor (e.g., Anthropic for Claude, OpenAI for GPT, DeepSeek, Alibaba for Qwen, Zhipu AI for GLM, Moonshot for Kimi).
- Model ID: The exact string to use in API requests. Always copy it from the list.
- Status: Indicates if the model is available, deprecated, or in preview.
- Context length and pricing: Shown per model. Pricing is official price × 1.3 for users; contributors receive official price × 1.1 or × 1.2 back in USDC.
How OpenRouter Exposes Models
OpenRouter provides a public model list and API. Their naming may include vendor prefixes (e.g., anthropic/claude-3-opus) and variants (e.g., :free). Availability and versioning are controlled by OpenRouter.
Comparing the Lists: Six Vendors Side by Side
For each vendor, compare the model IDs, versions, and statuses between our list and OpenRouter's.
- Claude (Anthropic): Check if the same Claude versions (e.g., Claude 3 Opus, Sonnet, Haiku) are present. Note any differences in naming or availability.
- GPT (OpenAI): Verify GPT-3.5, GPT-4, and newer variants. OpenRouter might expose additional preview models.
- DeepSeek: Compare DeepSeek models (e.g., DeepSeek-V2, DeepSeek-Coder). Ensure the exact ID matches.
- Qwen (Alibaba): Qwen models may have multiple sizes (e.g., Qwen-72B, Qwen-1.8B). Check which sizes are available on each platform.
- GLM (Zhipu AI): GLM-4 and other versions. OpenRouter may list them under different names.
- Kimi (Moonshot): Kimi models, often with long context. Confirm ID and context length.
How to Pick the Right Model ID
- Decide which vendor and model family you need.
- Look up the model in our list.
- If it exists, copy the exact model ID.
- If it doesn't, check OpenRouter's list for an equivalent, but note that you cannot call OpenRouter models through our API. You would need a separate account.
- When in doubt, use the model ID from our list, as that is what our API accepts.
Handling Model Updates
Model lists change as vendors release new versions or deprecate old ones. Subscribe to our changelog or API for updates. OpenRouter may update at a different pace, so always re-check before relying on a model ID.
Cost and Access
We offer a single API key for all supported models. Top up with USDC on Base, no KYC. You pay official price × 1.3. If you contribute keys, you are credited official price × 1.1 (premium × 1.2) in USDC. This is not a discount on official price; it's a simple, transparent markup and reward system.
Summary
Our model list and OpenRouter's are separate. Use our list to find the exact model ID for Claude, GPT, DeepSeek, Qwen, GLM, or Kimi. Compare with OpenRouter only to understand differences, not to assume compatibility. Always copy the ID from our list before making API calls.