Unquant

Opening your workspace…

Unquant

GLM-5.3-Flash

Our 328B sparse MoE flagship, uncensored at the tensor level: frontier-class reasoning for complex coding, long-running agents and deep analysis, with vision, tool calling and the full 262K context on every tier.

Our most capable model, for complex coding, long-running agents, deep reasoning and analysis, with the full 262K context on every tier. Works with any client that accepts a custom OpenAI-compatible base URL and model ID.

At a glance

  • Model family: GLM-5.3-Flash
  • Model ID: glm-5.3-flash
  • Plans from $109 / month
  • Usage: unlimited tokens & messages
  • Context: 262,144 tokens on every plan
  • Uncensored at the tensor level, with vision and tool calling
  • Access: private beta · by invitation

GLM-5.3-Flash subscriptions

  • GLM Standard: $109/month · unlimited tokens & messages · 262K context · 1 concurrent stream
  • GLM Pro: $199/month · unlimited tokens & messages · 262K context · 2 concurrent streams
  • GLM Ultra: $399/month · unlimited tokens & messages · 262K context · 4 concurrent streams
  • GLM Dedicated: $5,999/month · unlimited tokens & messages · 262K context · 8 concurrent streams

Compare plans · Request beta access · Read the docs

What GLM-5.3-Flash is built for

  • Complex coding: Multi-file changes, refactors and debugging with the whole repository in context.
  • Agents: Tool calling and adjustable reasoning for long, multi-step runs in Claude Code, Codex or OpenCode.
  • Analysis: Contracts, research and long documents, read in full.
  • Serious fiction: Plot-heavy long-form writing that stays consistent over hundreds of messages.

Specifications

  • Parameters: 328B (Sparse mixture of experts)
  • Context: 262K (Tokens, on every plan)
  • Input: Text + images (Vision built in)
  • Reasoning: 4 levels (none · low · high · max)
  • Model ID: glm-5.3-flash
  • Base model: GLM-5.3-Flash by Z.ai, open weights
  • Output: Text, with reasoning returned separately
  • API: OpenAI and Anthropic compatible
  • Tool calling, Vision, Streaming, JSON output, OpenAI & Anthropic APIs, Uncensored in the weights, Unlimited tokens

Connect in a minute

Any OpenAI-compatible client works. In Python:

from openai import OpenAI

client = OpenAI(base_url="https://api.unquant.io/v1", api_key="YOUR_UNQUANT_KEY")
reply = client.chat.completions.create(
    model="glm-5.3-flash",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(reply.choices[0].message.content)

Setup guides: SillyTavern · Janitor AI · Open WebUI · Claude Code · API docs

GLM-5.3-Flash FAQ

Is GLM-5.3-Flash really uncensored?

Yes. Refusal behaviour is removed in the model weights (tensor-level uncensoring), so there is no jailbreak prompt to maintain. Vision, tool calling and reasoning stay intact.

Is it really unlimited?

Yes. Every GLM-5.3-Flash plan includes unlimited tokens and messages with no caps or meters. Plans differ only in how many requests can run at once.

How much context do I get?

The full 262,144-token window on every plan, with no lower per-request cap.

Can I turn the reasoning off?

Yes. Send reasoning_effort "none" for direct answers, or "low", "high" (the default) or "max".