Dobby is freeeeeee — no key, no signup, no card. We don't read docs, we call the endpoint and publish what came back, including the 402s.

uncloseai

uncloseai (unturf) — docs

Call it

Base URL
https://hermes.ai.unturf.com/v1
API key
none — send no Authorization header at all
curl https://hermes.ai.unturf.com/v1/chat/completions \
  -H 'Content-Type: application/json' \
  -d '{"model":"Lorbus/Qwen3.6-27B-int4-AutoRound","messages":[{"role":"user","content":"hello"}]}'
OPENAI_BASE_URL=https://hermes.ai.unturf.com/v1
OPENAI_API_KEY=not-needed
OPENAI_MODEL=Lorbus/Qwen3.6-27B-int4-AutoRound

Using Lorbus/Qwen3.6-27B-int4-AutoRound, answered 102 of the 152 checks we made in the last week, last answered 32 h ago from this server's address. Every model this endpoint offers is in the table below.

Claimed free forever with no signup, self-published and carried by OmniRoute's routing registry as an endpoint needing no key. Every call from here on 2026-08-11 -- listing and chat, repeated -- answered 502, so the claim is seeded and the probe log is the answer about whether it comes back. Its terms throttle by IP for heavy use and forbid building a competing inference service on it; an hourly liveness call is neither, and the budget above is set low to keep it that way.

Models

ModelTierKeylessChatToolsSchemaJSONVisionDrawsWorkedLast checkedLast listed
Lorbus/Qwen3.6-27B-int4-AutoRoundyesyesyesyesyesyes 102/15221 min ago32 h ago
RedHatAI/Qwen3.8-27B-INT4yesyesyesyesyesyes 4/161 h ago22 h ago

Scroll the table sideways for the rest of each row. Chat, Worked and Last listed are folded away at this width.

Feature columns are measurements. Where the provider's own listing claims a feature we could not get out of it, the cell says so — that disagreement is the reason this shelf exists.

What Tools, Schema, JSON and Vision mean, and what each cell state means

The columns

Each name links to the call this endpoint's verdict was read from.

Tools
We sent a tools array and the model came back with a well-formed tool_calls entry an agent could dispatch. A 200 with no tool call in it is not tool support, and is recorded as a no.
Schema
The strict structured-output mode: response_format: {"type": "json_schema", …}. The reply parsed and matched the schema we sent.
JSON
The weaker mode: response_format: {"type": "json_object"}. Valid JSON, with no shape enforced. The two come apart in both directions — OVH's Mistral-7B-Instruct-v0.3 does the strict form and emits invalid JSON for the weak one; llm7's meta-Llama-3.1-8B-Instruct-Turbo does json_object while json_schema answers 405.
Vision
A real image went in inline and a correct answer about it came back. We ask about something only a model that looked could know.
Draws
On an endpoint's own page: bytes came back with an image content type, and they really are an image. Not a 200 taken on trust.

The cell states

yes
A real call proved it, at the time the endpoint page prints beside that call. The dot's tooltip says the same thing where there is a pointer to hover with.
no
A clean negative: the parameter was refused, or the answer did not conform.
no, claimed
The negative contradicts the provider's own published claim — their listing advertises the feature and our call could not get it out of the model. This is the most interesting cell on the site and the reason the shelf exists.
never probed
No verdict. Either nobody has asked yet, or the check ran and got no usable answer back — Pollinations sits here because its anonymous pool answered 402 to every feature probe. It is a question, never a no.

The same columns, in the API

curl 'https://stillworks.supercapybara.com/api/llm/up?feature=tools,json_schema'

The names are the API's spelling of the columns: tools, json_schema, json_object, vision, image_out. Several are and, never or — a row comes back only if a real call proved every feature named. Models we never probed for one of them are left out rather than optimistically included, so a filtered shelf is shorter than a hopeful one.

The call behind each verdict

One row per feature: the most recent call that decided that column, whatever it answered. A refusal is the evidence behind a no exactly as a success is the evidence behind a yes, so both are here.

FeatureWhenOutcomeHTTPLatencyModelDetail
Tools1 h agoserver_error502300 msLorbus/Qwen3.6-27B-int4-AutoRound
Schema1 h agoserver_error502297 msLorbus/Qwen3.6-27B-int4-AutoRound
JSON1 h agoserver_error502358 msLorbus/Qwen3.6-27B-int4-AutoRound
Vision1 h agoserver_error502382 msLorbus/Qwen3.6-27B-int4-AutoRound

Raw probe log

Our working, so you can check it rather than trust it. Wider than the page at any width — scroll it sideways for the Detail column, which quotes the provider's own refusal back.

WhenKindOutcomeHTTPLatencyModelDetail
21 min agochatserver_error502284 msLorbus/Qwen3.6-27B-int4-AutoRound
21 min agomodelsserver_error502994 ms
1 h agochatserver_error502296 msRedHatAI/Qwen3.8-27B-INT4
1 h agomodelsserver_error5021022 ms
1 h agovisionserver_error502382 msLorbus/Qwen3.6-27B-int4-AutoRound
1 h agojson_objectserver_error502358 msLorbus/Qwen3.6-27B-int4-AutoRound
1 h agojson_schemaserver_error502297 msLorbus/Qwen3.6-27B-int4-AutoRound
1 h agotoolsserver_error502300 msLorbus/Qwen3.6-27B-int4-AutoRound
1 h agovisionserver_error502333 msRedHatAI/Qwen3.8-27B-INT4
1 h agojson_objectserver_error502312 msRedHatAI/Qwen3.8-27B-INT4
1 h agojson_schemaserver_error502307 msRedHatAI/Qwen3.8-27B-INT4
1 h agotoolsserver_error502765 msRedHatAI/Qwen3.8-27B-INT4
2 h agochatserver_error5025213 msLorbus/Qwen3.6-27B-int4-AutoRound
2 h agomodelsserver_error5025831 ms
3 h agochatserver_error5025182 msRedHatAI/Qwen3.8-27B-INT4
3 h agomodelsserver_error5026075 ms
4 h agochatserver_error5025231 msLorbus/Qwen3.6-27B-int4-AutoRound
4 h agomodelsserver_error5025833 ms
5 h agochatserver_error502316 msRedHatAI/Qwen3.8-27B-INT4
5 h agomodelsserver_error5021169 ms
6 h agochatserver_error502295 msLorbus/Qwen3.6-27B-int4-AutoRound
6 h agomodelsserver_error5021284 ms
7 h agochatserver_error502328 msRedHatAI/Qwen3.8-27B-INT4
7 h agomodelsserver_error502925 ms
8 h agochattimeout25000 msLorbus/Qwen3.6-27B-int4-AutoRoundPost "https://hermes.ai.unturf.com/v1/chat/completions": context deadline exceeded (Client…
8 h agomodelstimeout25000 msGet "https://hermes.ai.unturf.com/v1/models": context deadline exceeded (Client.Timeout ex…
9 h agochatserver_error502317 msRedHatAI/Qwen3.8-27B-INT4
9 h agomodelsserver_error502857 ms
10 h agochatserver_error502316 msLorbus/Qwen3.6-27B-int4-AutoRound
10 h agomodelsserver_error502989 ms
11 h agochatserver_error502463 msRedHatAI/Qwen3.8-27B-INT4
11 h agomodelsserver_error502905 ms
12 h agochatserver_error502330 msLorbus/Qwen3.6-27B-int4-AutoRound
12 h agomodelsserver_error502940 ms
13 h agochattimeout25001 msRedHatAI/Qwen3.8-27B-INT4Post "https://hermes.ai.unturf.com/v1/chat/completions": context deadline exceeded (Client…
13 h agomodelstimeout25000 msGet "https://hermes.ai.unturf.com/v1/models": context deadline exceeded (Client.Timeout ex…
14 h agochatserver_error502284 msLorbus/Qwen3.6-27B-int4-AutoRound
14 h agomodelsserver_error502739 ms
15 h agochatserver_error502295 msRedHatAI/Qwen3.8-27B-INT4
15 h agomodelsserver_error502884 ms