Z.ai
Chat with GLM 4.7 online
Z.ai's fast open-weight coder — reasons before every tool call and keeps that reasoning across the whole job.
200K tokens
Context
8,192 tokens
Max output
Fast
Speed
Straight from GLM 4.7
Three prompts, three unedited answers, generated on September 2, 2026 through NinjaChat. Click the prompt to copy it and run it yourself.
Write a Python function that merges overlapping intervals, plus three short tests. Keep the whole answer under 40 lines of code.
def merge_intervals(intervals):
if not intervals:
return []
intervals.sort(key=lambda x: x[0])
merged = [intervals[0]]
for current in intervals[1:]:
last = merged[-1]
if current[0] <= last[1]:
last[1] = max(last[1], current[1])
else:
merged.append(current)
return merged
# Tests
assert merge_intervals([[1,3],[2,6],[8,10],[15,18]]) == [[1,6],[8,10],[15,18]]
assert merge_intervals([[1,4],[4,5]]) == [[1,5]]
assert merge_intervals([]) == []
Read: We ran the same 3 prompts on 12 AI models — Unedited answers, latency and price for every model, measured through the NinjaChat API in September 2026.
Served by DeepInfra as glm-4.7. Answers vary run to run; these are one take each, not the best of several.
What you’ll use it for
Everyday coding
Fast, accurate code generation and debugging across major languages.
Tool-using agents
Reasoning before every tool call, kept across the whole task.
Long sessions
Preserved thinking keeps a multi-step job consistent from start to finish.
Math and logic
Strong on olympiad-style math and structured reasoning.
Prompts to steal
Tuned to this model — click any line to copy.
About GLM 4.7
GLM 4.7 is Z.ai's open-weight foundation model, released on December 22, 2025 and built for real development work: long task cycles, frequent tool use, and code that has to be right rather than plausible. On NinjaChat it is included in every plan, next to Claude, GPT-5.6 and Gemini, with no Z.ai account.
What makes 4.7 distinctive is how it thinks. Z.ai gave it three reasoning modes that you feel rather than configure: it reasons before each response and each tool call, it keeps that reasoning across turns so a long session stays consistent, and it can turn reasoning up or down per turn so quick questions stay quick. The result is a model that behaves steadily across a multi-step job instead of drifting.
It is a genuinely strong coder. Z.ai reports 73.8% on SWE-bench Verified and 87.4% on τ²-Bench, the interactive tool-use benchmark, where it posted the highest score of any open model at release, alongside 95.7% on AIME 2025 for math. In use that shows up as fewer hallucinated APIs, cleaner diffs, and better behavior when it has to call tools and read what came back.
GLM 4.7 handles about 200K tokens of context on NinjaChat's rail, enough for a sizeable codebase or a long document set, and it is fast, which is why many people keep it as their everyday model and escalate to a heavier one only when needed. GLM 5 and GLM 5.2 sit above it in the family; 4.7 is the pragmatic middle, and its weights are open if you ever want to run it yourself.
If you have been searching for a GLM chat you can just open and use, this is it: no waitlist, no separate app, the same account and plan as every other model here.
How to use GLM 4.7
From zero to your first result in under a minute.
01
Create a NinjaChat account and choose a plan
02
Open chat and pick GLM 4.7 from the model list
03
Ask for code, a refactor or a step-by-step analysis
04
Keep the task in one thread; it carries its reasoning across turns
05
Escalate to GLM 5.2 or Claude when a problem needs more depth
Tips for better results
- Give it the error output verbatim; it debugs from evidence
- Ask it to call tools or search when the answer depends on facts
- It is fast: iterate in small steps rather than one giant prompt
- For heavy reasoning, tell it to think longer on that turn
GLM 4.7 vs the alternatives
Where it wins, and where it doesn't.
vs GLM 5.2
ViewFaster and cheaper for everyday coding and chat
GLM 5.2 has more depth on systems design and long reasoning
vs Claude Sonnet 5
ViewOpen weights and strong tool-use scores at lower cost
Sonnet 5 has more polish on writing and broader tool ecosystems
vs DeepSeek V4
ViewPreserved thinking across turns and top open-model tool-use results
DeepSeek V4 offers a larger context window and different pricing
Questions, answered.
Beyond chat — images & video
The same subscription covers the image studio and video generation.
GLM 4.7, and 50 more.
One subscription covers every model on NinjaChat, GLM 4.7 included.
