GLM-5.2 Free Bartering Guide: SenseNova Daily New Token Plan, Use Flagship Model for Free

The Chinese AI community is experiencing a strange phenomenon: the stronger the model, the fewer free channels are available.

Just before the launch of Zhipu's GLM-5.2, Coding Plan users surged, with API-to-token priced at $1.4 USD per million tokens—affordable for individual developers, though certainly not "trivial delivery." As open weights emerged, local deployment required GPUs. The window for zero-cost flagship model experience was brief.

But SenseNova's Token Plan has actually opened this window into a manageable door.

During Public Beta, GLM-5.2 is completely free. This is not a superficial "register and get 1000 tokens" offer. It offers a quota of 1500 calls per 5 hours that continuously refreshes and remains free. Moreover, it extends beyond GLM-5.2: SenseNova 6.7 Flash-Lite, SenseNova U1 Fast, and DeepSeek V4 Flash are all included in this free plan.

01 What Does the SenseNova Token Plan Actually Provide?

The main takeaway: Free package, 0 cost, currently in public beta.

Note that DeepSeek V4 Flash only offers 150 calls—SenseNova acts as a reseller, charging DeepSeek for each conversion, making 150 calls akin to a "try-before-you-buy" ticket. However, based on actual results, the text generation speed reaches 140 tokens/s, performing exceptionally well in the free options.

By contrast, GLM-5.2's 1500 calls per 5 hours constitute a true buffet.

02 Why GLM-5.2 Merits Free Use

On June 13th, Zhipu has opened GLM Coding Plan users to access GLM-5.2. The core parameters of this model are:

• 1M Context Window — Currently the longest among domestic models; there is still room for a complete book of Three Body Problem.
• High/Max Double Thought Intensity Levels — For code tasks, Max intensity is recommended for deeper reasoning.
• MIT Open Source License — Weights can be freely used, including for commercial purposes.
• API Pricing: Input 1.4 USD/Million Tokens, Output 4.4 USD/Million Tokens — About 1/7 of the rate of Claude.

All these capabilities are completely free under the SenseNova Token Plan. You do not need to purchase the Coding Plan, do not need to top up your API balance, and registration alone provides access.

03 How to Integrate: Complete in 3 Minutes

Step 1: Register

Visit https://www.sensenova.cn/token-plan and register an account, selecting the Free plan.

Step 2: Obtain the API Key

Enter the console and create an API Key. A maximum of 20 keys can be created.

Step 3: Integrate Your Tools

SenseNova's API is compatible with the OpenAI format. The API endpoint is:

https://token.sensenova.cn/v1/chat/completions

Integrate with Claude Code:

export OPENAI_API_KEY="Your SenseNova Token API Key"
export OPENAI_BASE_URL="https://token.sensenova.cn/v1"

Integrate with Hermes Agent:

Add the custom provider to the config.yaml:

custom_providers:
  - name: sensenova
    base_url: https://token.sensenova.cn/v1
    key_env: SENSENOVA_API_KEY

Integrate with OpenClaw:

SenseNova officially supports OpenClaw and Hermes Agent rapid integration. A one-click configuration entry is available on the Token Plan page.

Direct Python Invocation:

from openai import OpenAI

client = OpenAI(
    api_key="Your SenseNova Token API Key",
    base_url="https://token.sensenova.cn/v1"
)

response = client.chat.completions.create(
    model="glm-5.2",
    messages=[{"role": "user", "content": "Write a quicksort in Python"}],
    stream=True
)

for chunk in response:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

04 Your Hide-and-Seek Card: More Than GLM

SenseNova's strategy is clear in its writing: Free tokens are the hook, Coworking Skills are the carp to catch.

SenseNova 6.7 Flash-Lite is not an ordinary chat model; it is a Multimodal Agent—capable of "seeing" webpage layouts, document structures, and financial charts directly, achieving an integrated "See, Think, Do" capability. SenseNova has encapsulated this into SenseNova-Skills, covering:

• 📊 Data Analysis — Automatically generates reports from 900K lines of sales data.
• 📑 PPT Generation — Transforms outlines into formatted layouts in a single click.
• 🖼 Infographic Creation — Specifically designed for U1 Fast, capable of generating infographics.
• 🔍 Deep Research — Produces vertically oriented industry reports with 8 chapters, authoring and arranging them independently.

These Skills are already open-sourced on GitHub and natively support Hermes Agent and OpenClaw frameworks.

05 Comparison with Other Free Channels

SenseNova's advantage lies in the true 0 yuan threshold, requiring no subscription regardless of level.

06 Several Things to Note

1. Public Beta ≠ Forever Free — SenseNova explicitly marked "Public Beta"; future Lite/Pro paid tiers may be launched. Take advantage of the free time now.
2. GLM-5.2's Multiplier Factor — On Zhipu's internal Coding Plan, consumption is 3x during peak hours (14:00-18:00) and 2x off-peak. On the SenseNova side, it is calculated at 1x, saving another layer.
3. Documentation Not Yet Updated — The SenseNova control center can see GLM-5.2 usage metrics, but official documentation for specific details is not yet in place; actual usage will proceed as the primary guide.
4. Domestic Cars Already Support — If using the Domestic Lobachat (Lobechat) client, GLM-5.2 can already be seen in the model list.

Final Words

By 2026, the AI market trend is becoming increasingly clear: The models themselves are depreciating, while the entryways and workflows are rising.

SenseNova offering free GLM-5.2 is not charity—it aims to get users to adapt to the SenseNova ecosystem, adapt to Coworking Skills workflows, and adapt to using SenseNova models in Hermes Agents and OpenClaw. Zhipu's logic is the same, sending 20 million tokens to new users.

However, for developers, this is the perfect window. As multiple manufacturers rush to market, free tokens abound; flagship models are easily accessed. Once a winning strategy is established, free meals disappear.

How to Obtain: https://www.sensenova.cn/token-plan

Promotion Time: Looking for more powerful models? Agnes AI offers 1M context window, 4K image generation, and video generation completely free; API address apihub.agnes-ai.com/v1 → https://platform.agnes-ai.com