GLM 5.3 Abliterated / Uncensored

orca/orcacyber-glm-5.3
FeaturedZDR
ToolsJSONReasoning
by OrcaCyber · 2026-11-04

Cybersecurity model for authorized security research, based on Z.ai GLM-5.3 with refusal behavior removed. Full offensive/defensive workflow: vulnerability research, exploit development, pentest/red-team tooling, auditing, malware analysis. Text in/out, 1M-token context, up to 128K output. Access-gated to the Security Research tier.

ctx1M tokens
Max output128K
Inputtext
Outputtext
p50 TTFT4.25 s
Data retentionZero Data Retention available. With ZDR on, requests go only to providers with a zero-retention agreement.
INPUT$3.00/ 1M tokens
OUTPUT$5.00/ 1M tokens
p50 TTFT4.25 s7d
p95 TTFT9.84 s7d
TRAFFIC5.8Ktokens / 7d
Get the GLM 5.3 Abliterated / Uncensored API →🔒 Enable access

Offensive-capability model · a passkey and the terms

A cybersecurity model based on Z.ai's GLM-5.3, purpose-built for authorized security research and security engineering across the full offensive and defensive workflow. Refusal behavior is removed,…

What is OrcaCyber GLM-5.3?

Pricing

Pricing
Input / 1M tokens$3.00
Output / 1M tokens$5.00
Cache read / 1M$0.300
CurrencyUSD

Cost calculator

Tokens / month10MM
Input share70%%
Estimated / month $36.00 · With prompt caching $26.55

Estimate based on list price

Token & cost estimator

Input tokens: 20Cost per request: $0.002560

Estimate only — actual token counts depend on the provider's tokenizer.

Performance

last 7 days
p50 TTFT
4.25 s
Output speed
38.0 tok/s
p95 TTFT
9.84 s
Error rate
0%

Public benchmarks

pending
How Design Arena works
Source: Design Arena

Community buzz

What developers are saying this week

Hacker News0 mentions · 7d

How it compares

GLM 5.3 Abliterated / UncensoredOrcaCyber Ultra 1.0OrcaCyber Zero 1.0OrcaCyber Zero 1.5
Input $/M$3.00$5.00$3.00$3.00
Output $/M$5.00$10.00$7.50$7.50
Context1.0M1.0M1.0M1.0M
Quality9/109/109/109/10
Compare side-by-sideCompare side-by-sideCompare side-by-sideCompare side-by-side

Browse AI model comparisons →

More from OrcaCyber

See all models from orca →
DeepSeek V4.1 Flash Abliterated / UncensoredNew
orca/orcacyber-flash-1.0
$0.80 in · $1.60 out / 1M
1M ctx· quality 8/10
 Compare side-by-side
GLM 5.3 Flash Abliterated / UncensoredNew
orca/orcacyber-glm-5.3-flash
$0.80 in · $1.60 out / 1M
1M ctx· quality 8/10
 Compare side-by-side
Hy4 Abliterated / Uncensored
orca/orcacyber-hy4
$0.80 in · $1.60 out / 1M
1M ctx· quality 8/10
 Compare side-by-side

FAQ

What is OrcaCyber GLM-5.3?
A cybersecurity model based on Z.ai's GLM-5.3, purpose-built for authorized security research and security engineering across the full offensive and defensive workflow. Refusal behavior is removed, so it engages directly with offensive-security tasks that general-purpose assistants decline.
Who can use it?
It is access-gated to the Security Research tier, intended for trusted security researchers, red teams, and authorized security testing.
What are the context and output limits?
A 1M-token context window — enough for whole repositories, large log sets, packet captures or disassembly listings in one call — with up to 128K output tokens for long single-pass artifacts.
What capabilities does it support?
Text in and out, thinking and non-thinking modes with selectable reasoning effort, native function calling, and structured (JSON) output, over an OpenAI-compatible API.

Embed this badge

Paste into your blog post
OrcaCyber: GLM 5.3 Abliterated / Uncensored•$3.00/M in•4250ms p50•via OrcaRouter
HTML <a href="https://www.orcarouter.ai/models/orca/orcacyber-glm-5.3" target="_blank"> <img src="https://www.orcarouter.ai/embed/orca/orcacyber-glm-5.3.svg" alt="OrcaCyber: GLM 5.3 Abliterated / Uncensored on OrcaRouter" /> </a>
Markdown [![OrcaCyber: GLM 5.3 Abliterated / Uncensored](https://www.orcarouter.ai/embed/orca/orcacyber-glm-5.3.svg)](https://www.orcarouter.ai/models/orca/orcacyber-glm-5.3)

Model card as data

GET /api/public/models/orca/orcacyber-glm-5.3Open
Machine-readable:/llms.txt/llms-full.txt