stillworks

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.

ovh-anonymous

OVHcloud AI Endpoints — docs

Call it

Base URL
https://oai.endpoints.kepler.ai.cloud.ovh.net/v1
API key
none — send no Authorization header at all
curl https://oai.endpoints.kepler.ai.cloud.ovh.net/v1/chat/completions \
  -H 'Content-Type: application/json' \
  -d '{"model":"Meta-Llama-3_3-70B-Instruct","messages":[{"role":"user","content":"hello"}]}'
OPENAI_BASE_URL=https://oai.endpoints.kepler.ai.cloud.ovh.net/v1
OPENAI_API_KEY=not-needed
OPENAI_MODEL=Meta-Llama-3_3-70B-Instruct

Using Meta-Llama-3_3-70B-Instruct, answered 1 of the 10 checks we made in the last week, last answered 5 days ago from this server's address. Every model this endpoint offers is in the table below.

Claimed permanent free anonymous tier, roughly 2 requests per minute per IP per model, no signup. Rate limit is low enough that a single probe per cycle is the ceiling of what is polite.

Models

ModelTierKeylessChatToolsSchemaJSONVisionDrawsWorkedLast checkedLast listed
Meta-Llama-3_3-70B-Instruct—yesyesyes — — — —1/109 h ago26 min ago
Mistral-7B-Instruct-v0.3—yesyesyesyesyesno —3/1010 h ago26 min ago
Mistral-Nemo-Instruct-2407—yesyesyesyesyesno —4/1015 h ago26 min ago
Mistral-Small-3.2-24B-Instruct-2506—yesyesyesyesyesyes —0/114 h ago26 min ago
Qwen2.5-VL-72B-Instruct—yesyesnoyesyesyes —2/116 h ago26 min ago
Qwen3-32B—yesyesyesyesyesno —2/1126 min ago29 h ago
Qwen3-Coder-30B-A3B-Instruct—yesyesyesyesyesno —1/1014 h ago26 min ago
Qwen3.5-397B-A17B—yesyesyesyes —yes —0/1011 h ago26 min ago
Qwen3.5-9B—yesyesyesyesyesyes —0/113 h ago26 min ago
Qwen3.6-27B—yesyesyesyesyesyes —2/111 h ago26 min ago
Qwen3.8-27B—yesyesyesyesyesyes —4/1012 h ago26 min ago
Qwen3Guard-Gen-0.6B—yesyesnononono —5/117 h ago26 min ago
Qwen3Guard-Gen-8B—yesyesnononono —7/108 h ago26 min ago
gpt-oss-120b—yesyesyesyesyesno —0/115 h ago26 min ago
gpt-oss-20b—yesyesyesyesyes — —4/112 h ago26 min ago
stabilityai/stable-diffusion-xl-base-1.0—yesnot a chat model — — — —yes—never44 days ago
stable-diffusion-xl-base-v10—yesnot a chat model — — — —yes—never26 min ago
Qwen3-Embedding-8B—not yet checkednot a chat model — — — — ——never26 min ago
bge-m3—not yet checkednot a chat model — — — — ——never26 min ago
bge-multilingual-gemma2—not yet checkednot a chat model — — — — ——never26 min ago
nvr-tts-de-de—not yet checkednot a chat model — — — — ——never26 min ago
nvr-tts-en-us—not yet checkednot a chat model — — — — ——never26 min ago
nvr-tts-es-es—not yet checkednot a chat model — — — — ——never26 min ago
nvr-tts-it-it—not yet checkednot a chat model — — — — ——never26 min ago
ppl—not yet checkedyes — — — — —0/1013 h ago44 days ago
whisper-large-v3—not yet checkednot a chat model — — — — ——never26 min ago
whisper-large-v3-turbo—not yet checkednot a chat model — — — — ——never26 min 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 agorate_limited429251 msMistral-Nemo-Instruct-2407API rate limit exceeded
Schema1 h agorate_limited429255 msMistral-Nemo-Instruct-2407API rate limit exceeded
JSON1 h agorate_limited429249 msMistral-7B-Instruct-v0.3API rate limit exceeded
Vision1 h agorate_limited429247 msMistral-7B-Instruct-v0.3API rate limit exceeded
Draws1 h agook20011113 msstable-diffusion-xl-base-v10

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
26 min agochatclient_error404256 msQwen3-32BThe model `Qwen3-32B` does not exist.
26 min agomodelsok2001721 ms—
1 h agochatrate_limited429248 msQwen3.6-27BAPI rate limit exceeded
1 h agomodelsok2001693 ms—
1 h agojson_schemarate_limited429255 msMistral-Nemo-Instruct-2407API rate limit exceeded
1 h agotoolsrate_limited429251 msMistral-Nemo-Instruct-2407API rate limit exceeded
1 h agovisionrate_limited429247 msMistral-7B-Instruct-v0.3API rate limit exceeded
1 h agojson_objectrate_limited429249 msMistral-7B-Instruct-v0.3API rate limit exceeded
1 h agojson_schemarate_limited429252 msMistral-7B-Instruct-v0.3API rate limit exceeded
1 h agotoolsok2005726 msMistral-7B-Instruct-v0.3
1 h agoimage_outok20011113 msstable-diffusion-xl-base-v10
1 h agoimage_outok20028669 msstabilityai/stable-diffusion-xl-base-1.0
2 h agochatrate_limited429257 msgpt-oss-20bAPI rate limit exceeded
2 h agomodelsok2002421 ms—
3 h agochatrate_limited429252 msQwen3.5-9BAPI rate limit exceeded
3 h agomodelsok2002430 ms—
4 h agochatrate_limited429366 msMistral-Small-3.2-24B-Instruct-2506API rate limit exceeded
4 h agomodelsok2001477 ms—
5 h agochatrate_limited429253 msgpt-oss-120bAPI rate limit exceeded
5 h agomodelsok2001683 ms—
6 h agochatrate_limited429256 msQwen2.5-VL-72B-InstructAPI rate limit exceeded
6 h agomodelsok2002189 ms—
7 h agochatrate_limited429259 msQwen3Guard-Gen-0.6BAPI rate limit exceeded
7 h agomodelsok2002204 ms—
8 h agochatrate_limited429248 msQwen3Guard-Gen-8BAPI rate limit exceeded
8 h agomodelsok200975 ms—
9 h agochatrate_limited429251 msMeta-Llama-3_3-70B-InstructAPI rate limit exceeded
9 h agomodelsok2001714 ms—
10 h agochatrate_limited429274 msMistral-7B-Instruct-v0.3API rate limit exceeded
10 h agomodelsok2002424 ms—
11 h agochatrate_limited429257 msQwen3.5-397B-A17BAPI rate limit exceeded
11 h agomodelsok2001689 ms—
12 h agochatrate_limited429262 msQwen3.8-27BAPI rate limit exceeded
12 h agomodelsok2001965 ms—
13 h agochatclient_error404242 mspplThe model `ppl` does not exist.
13 h agomodelsok2001949 ms—
14 h agochatrate_limited429250 msQwen3-Coder-30B-A3B-InstructAPI rate limit exceeded
14 h agomodelsok2001693 ms—
15 h agochatrate_limited429247 msMistral-Nemo-Instruct-2407API rate limit exceeded
15 h agomodelsok2001925 ms—