Provider
OpenRouter API: models, pricing and compatibility
A router in front of many upstream inference providers. It lists the standard (not refusal-reduced) versions of most base models on this site, plus Venice Uncensored, and routes each request to one of several upstream hosts.
At a glance
- Type
- Router in front of other providers
- API base URL
- https://openrouter.ai/api/v1Source: Quickstart
- Billing
- Prepaid credit, charged per token at the price of the upstream provider that served the request.
- Docs
- https://openrouter.ai/docs/quickstart
- Prices
- https://openrouter.ai/models
- Commercial relationship with Atlas
- None.
Documented features
| Feature | Status per provider docs | Notes |
|---|---|---|
| OpenAI-style chat completions | Supported | The OpenAI SDK can be pointed at OpenRouter. |
| Public model list | Supported | Public GET /api/v1/models with prices and supported parameters. |
| Streaming (SSE) | Supported | |
| Tool / function calling | Per model or partial | Listed per model in supported_parameters. |
| JSON mode / response_format | Per model or partial | Listed per model (response_format, structured_outputs). |
| Image input | Per model or partial | Depends on the model. |
What to know
- The price in OpenRouter's model list is a single figure, but each model is served by several upstream providers at different prices and quantizations. Check the model's provider list before relying on a rate.
- Most OpenRouter listings for these base models are the standard instruct weights, so they are not a like-for-like substitute for a refusal-reduced variant.
Refusal-reduced offerings
| Provider and model id | What is served | Input / 1M | Output / 1M | Cached input / 1M | Context | Max output | 100K in + 20K out | Checked |
|---|---|---|---|---|---|---|---|---|
| OpenRouter cognitivecomputations/dolphin-mistral-24b-venice-edition | Names the exact Hugging Face repo | $0.20 | $0.90 | not listed | 128,000 | Not stated | $0.0380 | 14 November 2026 |
Standard models in our data
| Provider and model id | What is served | Input / 1M | Output / 1M | Cached input / 1M | Context | Max output | 100K in + 20K out | Checked |
|---|---|---|---|---|---|---|---|---|
| OpenRouter z-ai/glm-5.3-flash | Standard model, not refusal-reduced | $0.0399 | $0.0784 | $0.03 | 1,048,576 | Not stated | $0.00556 | 14 November 2026 |
| OpenRouter google/gemma-4-26b-a4b-it | Standard model, not refusal-reduced | $0.0675 | $0.225 | $0.0375 | 262,144 | Not stated | $0.0113 | 14 November 2026 |
| OpenRouter z-ai/glm-4.7-flash | Standard model, not refusal-reduced | $0.0605 | $0.40 | not listed | 200,000 | Not stated | $0.0140 | 14 November 2026 |
| OpenRouter qwen/qwen3.8-27b | Standard model, not refusal-reduced | $0.15 | $1.875 | $0.0375 | 1,024,000 | Not stated | $0.0525 | 14 November 2026 |
| OpenRouter meta-llama/llama-3.3-70b-instruct | Standard model, not refusal-reduced | $0.71 | $0.71 | not listed | 131,072 | Not stated | $0.0852 | 14 November 2026 |
google/gemma-4-26b-a4b-it: Upstream rates in OpenRouter's provider list on 2026-11-14: input $0.023 to $0.14, output $0.07 to $0.40 per 1M across active routes.meta-llama/llama-3.3-70b-instruct: Upstream rates in OpenRouter's provider list on 2026-11-14: input $0.10 to $0.72, output $0.32 to $0.79 per 1M across active routes.qwen/qwen3.8-27b: Upstream rates in OpenRouter's provider list on 2026-11-14: input $0.035 to $0.42, output $1.05 to $3.00 per 1M across routes.z-ai/glm-4.7-flash: Upstream rates in OpenRouter's provider list on 2026-11-14: one active route at $0.0605 input and $0.40 output per 1M.z-ai/glm-5.3-flash: Upstream rates in OpenRouter's provider list on 2026-11-14: input $0.0399 to $0.225, output $0.0784 to $0.75 per 1M across about 40 routes.
Sources
- OpenRouter model list (/api/v1/models), OpenRouter. Checked 14 November 2026.
- Quickstart, OpenRouter. Checked 14 November 2026.
- List available models (API reference), OpenRouter. Checked 14 November 2026.