Free · 50 million tokens · just a work email

The most cyber‑capable open model, for defenders

Point Claude Code, Codex or opencode at one URL. Free API key with 50 million tokens — confirm your work email and go.

base_url

Get a free API key See what it does

Your free allowance

50,000,000 free tokens

Per confirmed work email. The squares empty as you use them.

Previous-generation models scored zero

Anthropic's Frontier Red Team measured whether models can find and exploit memory-safety bugs in open-source software and land a full control-flow hijack. GLM-5.3 crosses a line that last generation's models never reached.

binary-exploitation.bench

Claude Mythos PreviewAnthropic · limited release

GLM-5.3vidoc-1 runs this

Claude Opus 4.6previous generation

GLM-5.2previous generation

0%2%4%6%8%

Full control-flow hijacks over 100 OSS-Fuzz tasks. On V8 end-to-end exploits (ExploitBench), GLM-5.3 succeeds 50 times in 410 attempts, close to Claude Mythos Preview's 56. Source: Anthropic Frontier Red Team.

"The most cyber-capable open-weight model released to date.": NIST CAISI assessment of GLM-5.3, Sep 2026
Builds working exploits end to end, on par with Claude Mythos Preview.: Anthropic Frontier Red Team, GLM-5.3 analysis

Security work it handles

Code review, vulnerability triage, authorized penetration tests, reverse engineering, malware analysis, threat modeling, CTF practice and report writing. A few examples of what a session looks like:

Application security

orders.py

> review this handler for vulnerabilitiesline 14 builds SQL from the URL path, soorder_id can change the query (SQL injection).- cur.execute(f"... WHERE id = {order_id}")+ cur.execute("... WHERE id = %s", (order_id,))▌

Penetration testing

staging-scan.txt

> rank what to check first on our staging host1  8080/tcp  admin panel over plain HTTP   confirm it should be reachable at all2  22/tcp    OpenSSH 7.4, years out of date   check vendor advisories, plan an upgrade3  443/tcp   TLS 1.0 still enabled   disable old protocol versions▌

Reverse engineering

sub_401a20.c

> what does this function do?It XORs each byte of a buffer with 0x5A andreturns the result: a string decoded at runtime, a common way to hide configuration inmalware. The decoded value is an update URL.renamed  sub_401a20 → decode_config_string▌

Security research

threat-model.md

> threat model our password reset flowhigh    reset tokens never expiremedium  token travels in a GET query string        and ends up in proxy logsmedium  no rate limit on /resetlow     reset email reveals if an account exists▌

These are illustrations of the kinds of tasks people use it for. Use it only on systems, code and binaries you own or are authorized to test.

Set up your agent

vidoc-1 speaks the OpenAI Chat Completions, OpenAI Responses and Anthropic Messages APIs, so each client only needs a new base URL, model name and API key. Get a key below, then follow the steps for your tool.

get-api-key

We email you a link; open it to get your key and 50 million tokens. Work email only: Gmail and disposable addresses aren't accepted. Lost your key? Sign up again with the same email to replace it.

Your telecommunications terminal equipment (e.g. telephone, computer) and automated calling systems, as defined by the Polish Electronic Communications Law of 12 July 2024, will be used to send commercial information. Detailed rules on how your personal data is processed: vidocsecurity.com/terms.

The model

What your agent gets on the other end of the URL.

vidoc-1.spec

base model: DeepSeek-V4.1-Flash-1017 (DeepSeek), with DeepSeek's own API conversion (deepseek-recipe) for the reasoning and tool-call format.
context: 262,144 tokens, shared by the prompt and the output. A response can use up to 32,768 tokens, or whatever room the prompt leaves if that's less.
thinking: Always on, at low effort by default to keep responses fast. For deeper reasoning, set the API's effort field to "high" or "max": reasoning_effort (Chat Completions), reasoning.effort (Responses) or output_config.effort (Messages).
tool calling: Supported in all three APIs.
endpoints: /v1/chat/completions, /v1/responses, /v1/messages, /v1/messages/count_tokens and /v1/models
api key: Required. Free with a confirmed work email; get one above. Send it as Authorization: Bearer or x-api-key.
model name: vidoc-1. Requests that name another model are served by vidoc-1 too.

Limits

Authorized use only. Test only systems, code and binaries you own or have permission to test.
50 million tokens per confirmed email, counting prompt and output tokens. Prompt tokens served from the model's cache count half. Gmail and disposable addresses can't sign up.
Coding agents resend the whole conversation on every turn, so a long session uses tokens quickly.
When your key runs out, requests return HTTP 429.
A stream you cancel is charged an estimate of what it used.
Request bodies can be at most 4.5 MB. Only the conversation, system prompt, tools, tool_choice, stream, the output limit and the reasoning effort are passed to the model. Other fields, including sampling settings such as temperature, are ignored.
Requests and responses are stored for security and compliance purposes. Don't send secrets or code you aren't allowed to share.

for-teams.txt

Want Mythos at home?

This free tier is a taste of frontier cyber capability. For unlimited, private use — deployed into your own environment, to your compliance requirements — we set it up for your team.

Book a call Enterprise security →