GLM-4.7 (Reasoning)
Intelligence
42.1
#71/532
Coding
36.3
#126/86
Math
95
#9/265
Speed
80 tok/s
TTFT: 713.00s
Pricing
$0.60 / $2.20
per 1M tokens (in/out)
GLM 4.7 is a coding‑centric model that thinks before acting, preserves its reasoning across turns, and lets you control thinking per request for speed or accuracy. It upgrades agentic workflows with stronger multi‑step tool use, better terminal and multilingual coding, and a noticeable jump in UI output quality for modern, clean webpages and slides. You can use it in popular coding agents, call it via the Z.ai API, and even run it locally with public weights on HuggingFace and ModelScope using vLLM or SGLang.
GLM-4.7 (Reasoning) by Z AI is designed for advanced reasoning tasks, achieving a Math Index of 95. It processes at a speed of 80.335 tokens per second and is priced at $0.6 per million input tokens, targeting professional users.
When to Use GLM-4.7 (Reasoning)
✓ Best For
- Complex mathematical problem solving.
- Data analysis requiring high reasoning capabilities.
- Applications in educational tools and tutoring systems.
✗ Not Ideal For
- Basic coding tasks due to a lower Coding Index of 36.3.
- Real-time applications where speed is critical, given the TTFT of 713 seconds.
How GLM-4.7 (Reasoning) Compares
Intelligence Index · Higher is better
Z AIAnthropicAlibaba
Benchmark Profile
Output Speed · tok/s
DeepSeekKwaiKATZ AIMistral
Math Index
GoogleOpenAIZ AIKimiKwaiKAT
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (25)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 42.1 | |
| Coding Index | 36.3 | |
| Math Index | 95 | |
| MMLU-Pro | 856% | |
| GPQA | 859% | |
| LiveCodeBench | 894% | |
| HLE | 251% | |
| SWE-Bench Verified | 73.8% | |
| SciCode | 45.1% | |
| IFBench | 67.9% | |
| LCR | 64% | |
| TerminalBench Hard | 31.8% | |
| Tau2 | 95.9% | |
| AIME 2025 | 95% | |
| BrowseComp | 52% | |
| BrowseComp-zh | 66.6% | |
| Humanity's Last Exam | 42.8% | |
| IMO-AnswerBench | 82% | |
| LiveCodeBench v6 | 84.9% | |
| SWE-bench Multilingual | 66.7% | |
| Tau-bench | 87.4% | |
| Terminal-Bench | 33.3% | |
| Terminal-Bench 2.0 | 41% | |
| Arena Chat | 21.7 | |
| Arena Coding | 10.6 |
Data: Artificial Analysis · Updated: April 10, 2026
Frequently Asked Questions (15)
When was GLM-4.7 (Reasoning) released?
GLM-4.7 (Reasoning) was released on December 22, 2025.
Who created GLM-4.7 (Reasoning)?
GLM-4.7 (Reasoning) was created by Z AI.
How intelligent is GLM-4.7 (Reasoning)?
GLM-4.7 (Reasoning) scores 42 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 27).
How fast is GLM-4.7 (Reasoning)?
GLM-4.7 (Reasoning) generates output at 72.5 tokens per second (based on the median across providers serving the model), which is above average compared to other open weight models of similar size (median: 57.7 t/s).
What is the latency of GLM-4.7 (Reasoning)?
GLM-4.7 (Reasoning) has a time to first token (TTFT) of 1.34s (based on the median across providers serving the model), which is very competitive compared to other open weight models of similar size (median: 2.17s).
How much does GLM-4.7 (Reasoning) cost?
GLM-4.7 (Reasoning) costs $0.60 per 1M input tokens (better than average, median: $0.60) and $2.20 per 1M output tokens (better than average, median: $2.20), based on the median across providers serving the model.
What is GLM-4.7 (Reasoning) API pricing?
GLM-4.7 (Reasoning) costs $0.60 per 1M input tokens and $2.20 per 1M output tokens (based on the median across providers serving the model). For a blended rate (3:1 input to output ratio), this is $1.00 per 1M tokens. Pricing may vary by provider.
How verbose is GLM-4.7 (Reasoning)?
When evaluated on the Intelligence Index, GLM-4.7 (Reasoning) generated 170M output tokens, which is at the higher end compared to other open weight models of similar size (median: 39M).
Is GLM-4.7 (Reasoning) a reasoning model?
Yes, GLM-4.7 (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
What input modalities does GLM-4.7 (Reasoning) support?
GLM-4.7 (Reasoning) supports text input.
What output modalities does GLM-4.7 (Reasoning) support?
GLM-4.7 (Reasoning) supports text output.
Can GLM-4.7 (Reasoning) process images?
No, GLM-4.7 (Reasoning) does not support image input. It can only process text.
Is GLM-4.7 (Reasoning) multimodal?
No, GLM-4.7 (Reasoning) is not multimodal. It only supports text input.
What is the context window of GLM-4.7 (Reasoning)?
GLM-4.7 (Reasoning) has a context window of 200k tokens. This determines how much text and conversation history the model can process in a single request.
Is GLM-4.7 (Reasoning) open source?
Yes, GLM-4.7 (Reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
