菜单
Query, filter, and select from OpenRouter's 400+ model catalog. Use when choosing models, comparing pricing, or checking capabilities. Triggers: 'openrouter models', 'list models', 'model catalog', 'compare models', 'available models'.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/jeremylongshore/claude-code-plugins-plus-skills --skill openrouter-model-catalog命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | openrouter-model-catalog |
| description | Query, filter, and select from OpenRouter's 400+ model catalog. Use when choosing models, comparing pricing, or checking capabilities. Triggers: 'openrouter models', 'list models', 'model catalog', 'compare models', 'available models'. |
| allowed-tools | Read, Write, Edit, Grep, Bash(python3:*), Bash(curl:*), Bash(jq:*) |
| version | 1.20.0 |
| license | MIT |
| author | Jeremy Longshore <jeremy@intentsolutions.io> |
| tags | ["saas","openrouter","models","catalog"] |
| compatibility | Designed for Claude Code, also compatible with Codex and OpenClaw |
Query the GET /api/v1/models endpoint to browse 400+ models, filter by capabilities, compare pricing, and check provider endpoints. No API key required for the models endpoint.
curl and jq for the command-line catalog queries — GET /api/v1/models itself requires no authOPENROUTER_API_KEY only for the Special Routers completion example — see the openrouter-install-auth skill for setuprequests for filtering, plus the OpenAI SDK for the openrouter/auto example (pip install requests openai)curl -s https://openrouter.ai/api/v1/models | jq '.data | length'; add ?supported_parameters=tools to filter to tool-calling models.pricing.prompt/pricing.completion are per token (multiply by 1M for readable rates), plus context_length, top_provider.max_completion_tokens, and architecture.modality.GET /api/v1/models/{id}/endpoints per List Providers for a Model.:free, :nitro, :floor, :extended, :thinking), or delegate selection entirely to openrouter/auto per Special Routers./api/v1/models — prices change frequently.# Full catalog (no auth required)
curl -s https://openrouter.ai/api/v1/models | jq '.data | length'
# → 400+
# Filter to text output models only
curl -s "https://openrouter.ai/api/v1/models?supported_parameters=tools" | jq '.data | length'
{
"id": "anthropic/claude-3.5-sonnet",
"name": "Claude 3.5 Sonnet",
"description": "Anthropic's most intelligent model...",
"context_length": 200000,
"pricing": {
"prompt": "0.000003",
"completion": "0.000015",
"image": "0.0048",
"request": "0"
},
"top_provider": {
"context_length": 200000,
"max_completion_tokens": 8192,
"is_moderated": false
},
"per_request_limits":
Key fields:
pricing.prompt / pricing.completion -- cost per token (not per million; multiply by 1M for readable rates)context_length -- max input tokenstop_provider.max_completion_tokens -- max output tokensarchitecture.modality -- text->text, text+image->text, etc.import requests
models = requests.get("https://openrouter.ai/api/v1/models").json()["data"]
# Find all free models
free_models = [m for m in models if m["pricing"]["prompt"] == "0"]
print(f"Free models: {len(free_models)}")
# Models with tool calling support
# (query with supported_parameters)
tool_models = requests.get(
"https://openrouter.ai/api/v1/models?supported_parameters=tools"
).json()["data"]
print(f"Tool-calling models: {len(tool_models)}")
# Sort by prompt price (cheapest first, excluding free)
paid = [m for m in models if float(m["pricing"]["prompt"]) > 0]
paid.sort(key=lambda m: float(m["pricing"]["prompt"]))
for m in paid[:10]:
cost_per_m = float(m["pricing"]["prompt"]) * 1_000_000
print(f" ${cost_per_m:.2f}/M tokens — {m['id']} ({m['context_length']//1000}K ctx)")
large_ctx = [m m models m[] >= ]
()
# See all providers and their pricing for a specific model
curl -s "https://openrouter.ai/api/v1/models/anthropic/claude-3.5-sonnet/endpoints" | jq '.data[] | {
provider: .provider_name,
price_prompt: .pricing.prompt,
price_completion: .pricing.completion,
context_length: .context_length,
quantization: .quantization
}'
Append a suffix to any model ID for variant behavior:
| Suffix | Effect | Example |
|---|---|---|
:free | Free tier (where available) | google/gemma-2-9b-it:free |
:nitro | Sort providers by throughput (faster) | anthropic/claude-3.5-sonnet:nitro |
:floor | Sort providers by price (cheapest) | openai/gpt-4o:floor |
:extended | Extended context window | anthropic/claude-3.5-sonnet:extended |
:thinking | Enable extended reasoning | anthropic/claude-3.5-sonnet:thinking |
| Model ID | Behavior |
|---|---|
openrouter/auto | Auto-selects best model for your prompt (powered by NotDiamond) |
openrouter/free | Routes to free models only |
# Let OpenRouter pick the best model
response = client.chat.completions.create(
model="openrouter/auto",
messages=[{"role": "user", "content": "Write a SQL query to find duplicate emails"}],
max_tokens=200,
)
print(f"Auto-selected: {response.model}") # Shows which model was chosen
| Model ID | Context | Cost (prompt/completion per 1M) |
|---|---|---|
google/gemma-2-9b-it:free | 8K | Free |
meta-llama/llama-3.1-8b-instruct | 128K | ~$0.06 / $0.06 |
anthropic/claude-3-haiku | 200K | $0.25 / $1.25 |
openai/gpt-4o-mini | 128K | $0.15 / $0.60 |
anthropic/claude-3.5-sonnet | 200K | $3.00 / $15.00 |
openai/gpt-4o | 128K | $2.50 / $10.00 |
openai/o1 | 200K | $15.00 / $60.00 |
Prices change frequently. Always verify via /api/v1/models.
id, context_length, pricing, top_provider, and architecture fields$0.06/M tokens — meta-llama/llama-3.1-8b-instruct (128K ctx)provider_name, prompt/completion pricing, context_length, quantizationopenrouter/auto requests, response.model reveals which model the router actually selectedFetch the catalog once, then slice it three ways with the filters from Python: Query and Filter:
models = requests.get("https://openrouter.ai/api/v1/models").json()["data"]
free = [m for m in models if m["pricing"]["prompt"] == "0"]
large = [m for m in models if m["context_length"] >= 128_000]
print(f"Total: {len(models)}, free: {len(free)}, 128K+: {len(large)}")
# Total: 267, free: 12, 128K+: 45 (counts drift as the catalog changes)
The same pass sorted by prompt price surfaces the cheapest paid options — meta-llama/llama-3-8b-instruct: $0.05/1M prompt tokens leads the list in the worked run. More worked examples: references/examples.md.
| Issue | Cause | Fix |
|---|---|---|
| Model ID not found at request time | Model renamed, removed, or typo | Re-query /api/v1/models; use exact ID from catalog |
| Stale pricing | Cached catalog data outdated | Refresh catalog hourly; pricing updates dynamically |
| Empty results with filter | No models match the filter criteria | Broaden the filter; check parameter spelling |
/api/v1/models for deprecation notices and new model additionssupported_parameters query filter to ensure models support features you need (tools, JSON mode, etc.)