Task Aliases
A task alias is the value you put in the model field to ask SpiderGate for a job instead of a specific model. SpiderGate resolves the alias to a pool of models and routes each request to the least busy one. The models behind an alias are not a ranked list: every one of them takes traffic.
{ "model": "spideriq/coding", "messages": [...] }The aliases come in three families. The live list, with the slots behind each alias and its recent usage, is always available at GET /api/gate/v1/aliases (no auth required):
curl "https://spideriq.ai/api/gate/v1/aliases"Treat that endpoint as the source of truth for which models sit behind an alias. This page names no model: the slots are edited as providers change, and a model named here would go stale.
Three families
**
spideriq/*— tuned for workers**: scrapers, batch enrichers, single-shot LLM calls. Biased toward fast, low-cost providers (Cerebras, Groq, NVIDIA NIM, Mistral) because cost dominates at batch scale.**
agent/*— tuned for live conversational agents**: chat interfaces, assistants, multi-turn sessions. Biased toward subscription primaries where multi-turn coherence and uptime beat per-token cost.**
opvs/*— tuned for the OPVS agent platform**, which splits its traffic by job size rather than by task: a cheap lane, a simple lane, and three heavier lanes for coding, planning and review.
Pick the family by what's calling: a background job → spideriq/*; a user-facing agent → agent/*.
spideriq/* — worker aliases
::table
Alias | Use it for
`spideriq/chat` | General conversation
`spideriq/classification` | Categorization tasks
`spideriq/coding` | Code generation, debugging, refactors
`spideriq/creative` | SEO copy and creative writing
`spideriq/design` | Design reasoning (site cloning)
`spideriq/design-vision` | Multimodal design reasoning (screenshots)
`spideriq/extraction` | Structured data extraction, JSON output
`spideriq/fast` | Lowest-latency simple tasks
`spideriq/free` | Zero-cost requests only
`spideriq/lead-analysis` | B2B lead analysis and scoring
`spideriq/opvs` | OPVS symbol-notation formatting
`spideriq/planning` | Strategic planning
`spideriq/research` | Long-context analysis and research
`spideriq/research-person` | Person research reports
`spideriq/summarization` | Text summarization
`spideriq/tool-use` | Function calling
`spideriq/translation` | Language translation
`spideriq/vision` | Image understanding, OCRagent/* — live-agent aliases
::table
Alias | Use it for
`agent/chat` | General customer-facing conversation
`agent/coding` | Code generation, refactors, debugging
`agent/creative` | Copywriting, brand voice, narrative
`agent/fast` | Quick utility, classification, short replies
`agent/live-chat` | Sustained multi-turn sessions where latency is felt turn by turn
`agent/planning` | Long-horizon planning, system design
`agent/research` | Long-context, multi-document synthesis
`agent/tool-use` | Heavy tool calling, multi-step, JSON
`agent/vision` | Image understanding for agentsopvs/* — OPVS platform aliases
::table
Alias | Use it for
`opvs/cheap` | High-volume OPVS calls where cost dominates
`opvs/coding` | OPVS code generation and refactors
`opvs/planner` | OPVS task planning and decomposition
`opvs/review` | OPVS code and output review
`opvs/simple` | Short single-shot OPVS utility callsNote:
spideriq/opvsis the one alias where SpiderGate injects a system prompt. It prepends OPVS notation rules so the model emits valid symbol output, and your own messages are appended after it.
Pick an alias in the dashboard
You do not have to send an alias from code to try one. The SpiderGate dashboard's Chat tab at app.spideriq.ai/dashboard/gate uses the same resolution path your agents do, so it is the quickest way to see what an alias actually serves.
The current alias is shown in the settings panel on the right, under MODEL / ALIAS:

Press ⌘J, or type /model in the composer, to open the model picker. An alias is marked with a TASK ALIAS badge and shows how it routes rather than naming a single model:

Concrete models are listed underneath, so you can compare an alias against pinning one model directly.
Tell an agent to use one
For agents, install the gateway skill and give the alias by name:
npx @opvs-ai/skills install @spideriq/gateway-skillsThen an instruction as plain as this is enough, because the skill carries the alias list and the call shape:
Use SpiderGate with the
spideriq/extractionalias to pull the contact fields out of these pages.
The skill's use-the-gateway surface covers chat completions, streaming and the alias list. See API Reference for the raw HTTP contract.
How resolution works
Resolve the alias to its slots. Each slot names a provider and a model.
Pool every slot whose provider has a usable key. A key marked unhealthy is left out.
Pick by least busy. Any slot in the pool can take the request. A slot's number is display order, not priority: a model you think of as "the fallback" takes its share of the traffic.
Retry. A failed attempt is retried up to 3 times, and a failing deployment rests for 60 seconds.
Hand over to another alias when the alias's own pool cannot serve. Only the aliases in the table below have one.
One exception to step 3 is a Codex slot. In slot 0 it is tried first, and the pool takes the request only if that attempt fails before any output. In any other slot it takes no traffic.
::table
Alias | Hands over to
`spideriq/coding` | `spideriq/chat`
`spideriq/extraction` | `spideriq/chat`
`spideriq/creative` | `spideriq/chat`
`spideriq/lead-analysis` | `spideriq/research`
`spideriq/research` | `spideriq/chat`
`spideriq/research-person` | `spideriq/research`
`spideriq/planning` | `spideriq/chat`
`spideriq/tool-use` | `spideriq/chat`
`spideriq/classification` | `spideriq/chat`
`spideriq/summarization` | `spideriq/research`
`spideriq/translation` | `spideriq/chat`
`spideriq/fast` | `spideriq/chat`
`spideriq/free` | `spideriq/chat`
`agent/coding` | `agent/chat`
`agent/tool-use` | `agent/chat`
`agent/planning` | `agent/chat`
`agent/creative` | `agent/chat`
`agent/research` | `agent/chat`
`agent/fast` | `agent/chat`
`agent/live-chat` | `agent/chat`
`agent/chat` | `agent/tool-use`
`opvs/planner` | `opvs/simple`
`opvs/coding` | `opvs/simple`
`opvs/review` | `opvs/simple`
`opvs/simple` | `opvs/cheap`Every other alias has no hand-over: when its own pool cannot serve, the request fails. That is spideriq/chat, spideriq/design, spideriq/opvs and opvs/cheap, and the three vision aliases spideriq/vision, spideriq/design-vision and agent/vision. The vision aliases have none on purpose: a pool of text models would answer about an image it never received.
X-SpiderGate-Fallback-From is sent when you asked for a task alias and the model that served is not the alias's slot-0 model. Its value is the slot-0 model. It is sent on streaming and non-streaming answers. Because an alias is a pool, the header does not mean that something failed: it means a slot other than slot 0 served. Add ?include_route_trace=true to a non-streaming request to see every attempt — see Chat Completions.
Aliases vs. direct models
You don't have to use an alias. You can pass a concrete model id (e.g. gpt-4o) to pin a single model — useful when you need exactly one model. The trade-off is that there is no pool behind it and no hand-over to another alias. For a model you picked from the catalog, the string to send is that row's call_as, not its id. See Models & Direct Routing.
Multi-modal endpoints (images, audio, embeddings) do not use aliases — you pass the real OpenAI model name. See Images, Audio & Embeddings.
Next steps
Send a request with your chosen alias — Chat Completions.
Inspect the live slots behind each alias at
GET /aliases— API Reference.Restrict which models or aliases a token may use — Agent Keys.