Apex Flash

The Apex Flash API

Wild West API serves Apex Flash under the model id apex-flash. It is GLM 5.3 Flash tuned for security work: pentesting, CTFs and malware analysis. It has 1.05M tokens of context, costs $0.30 per million input tokens and $1.00 per million output tokens, with cached input at $0.15, and supports tool calling and reads images. One key works on the OpenAI-compatible endpoint (https://wildwestapi.com/v1) and on the Anthropic-compatible one Claude Code uses.

It always reasons before it answers. A request that asks it not to still works, and the answer comes back with its reasoning, which is billed as output.

Security-tuned, not uncensored

Apex Flash has not had its refusals taken out, so it carries no uncensored tag and it can decline. If it turns down a task you need done, Hipshooter, the uncensored GLM 5.3 Flash, is a separate option.

Apex Flash or GLM 5.3 Flash Cyber?

Apex Flash is tuned for pentesting, CTFs and malware analysis, and GLM 5.3 Flash Cyber for code review, exploit analysis and incident response. Both are priced the same. They take the same calls, so changing one string runs the same prompt on the other, which is an easy way to compare drafts. See GLM 5.3 Flash Cyber.

Playbooks to start from

Each one has a real prompt, a real call, and what to check before you trust the output.

All twelve are on the cybersecurity playbooks page.

Apex Flash against the rest of the list

Prices are per million tokens. The same list answers at GET https://wildwestapi.com/v1/models.

Call it with curl

curl https://wildwestapi.com/v1/chat/completions \
  -H "Authorization: Bearer sk-ww-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "apex-flash",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Call it with the OpenAI SDK

from openai import OpenAI

client = OpenAI(base_url="https://wildwestapi.com/v1", api_key="sk-ww-...")

resp = client.chat.completions.create(
    model="apex-flash",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)

In Claude Code

Claude Code adds /v1/messages itself, so its base URL has no path. The Claude Code setup guide has the rest.

export ANTHROPIC_BASE_URL=https://wildwestapi.com
export ANTHROPIC_AUTH_TOKEN=sk-ww-...
export ANTHROPIC_MODEL=apex-flash

FAQ