2026/11/18The November 2026 edition ranks 6 APIs and 8 models
Z.ai

GLM-5.3 Flash Abliterated

GLM-5.3 Flash Abliterated is Z.ai's 320B mixture-of-experts model with its refusal direction removed, built for high-volume work.

Rank#4 / 8
#4Editors' ranking
Size#4 / 8
320BTotal, 18B active
Price#2 / 5
$0.43Blended per 1M, lowest host
Context
1MTokens per request

Our pick for high-volume work at the lowest input price.

Input
Price in / out$0.24 / $1.00
Context1M tokens
Max output131,072
Model idglm-5.3-flash-uncensored

Its input price matches DeepSeek V4.1 Flash, and its output costs $1.00 per 1M tokens against $0.90.

Where to run GLM-5.3 Flash by API

ProviderModel idInputCachedOutputContextPrompts
Silk Computeglm-5.3-flash-uncensored$0.24$0.06$1.001MNever stored

Per 1M tokens, read from pricing pages on November 18, 2026.

Try GLM-5.3 Flash in two lines

Set the base URL and your key in the OpenAI or Anthropic SDK.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.silkcompute.com",
    api_key="YOUR_API_KEY",
)
reply = client.chat.completions.create(
    model="glm-5.3-flash-uncensored",
    messages=[{"role": "user", "content": "..."}],
)

In Claude Code:

export ANTHROPIC_BASE_URL=https://api.silkcompute.com
export ANTHROPIC_AUTH_TOKEN=YOUR_API_KEY
export ANTHROPIC_MODEL=glm-5.3-flash-uncensored
claude
GLM-5.3 Flash on Silk Compute

Compared with

Other abliterated models