Cheaper overall
GLM 4.7 Flash
100% lower blended rate
Higher benchmark scores
GLM 4.7
leads on 15 of 17 figures
Bigger context window
GLM 4.7
204,800 vs 200,000 tokens
Cheaper cached input
GLM 4.7 Flash
Free vs $0.11, 100% less
Where each one wins
GLM 4.7 Flash
- Cheaper input tokens Free vs $0.60, 100% less
- Cheaper output tokens Free vs $2.20, 100% less
- Cheaper cached input Free vs $0.11, 100% less
- Cheaper on all four workloads
- Higher support-agent performance (ฯยฒ-Bench Telecom) 98.8 vs 95.9
- Higher Uicomponent Elo 1,209 vs 1,201
GLM 4.7
- Larger context window 204,800 tokens
- Higher overall intelligence (Intelligence Index) 22.2 vs 14.9
- Higher graduate-level science (GPQA Diamond) 85.9 vs 58.1
- Higher expert-exam performance (Humanity's Last Exam) 27.4 vs 7.6
- Ahead on 12 more benchmark figures
What four workloads cost
One run, and the same run a thousand times.
Caching is what decides it: over twenty calls GLM 4.7 Flash costs $0.0000 against $0.1208, a difference of $120.80 across a thousand loops.
| Workload | GLM 4.7 Flash | GLM 4.7 | ร1,000 | Difference |
|---|---|---|---|---|
| Chat turn 1,000 in / 500 out | $0.0000 | $0.0017 | $0.0000 / $1.70 | GLM 4.7 Flash is 100% cheaper |
| RAG answer 10,000 in / 800 out | $0.0000 | $0.0078 | $0.0000 / $7.76 | GLM 4.7 Flash is 100% cheaper |
| Agent loop 32,000 in / 700 out ร 20 calls | $0.0000 | $0.1208 | $0.0000 / $120.80 | GLM 4.7 Flash is 100% cheaper |
| Long document 400,000 in / 3,000 out | GLM 4.7 Flash's context window holds 200,000 tokens, so a 400,000-token prompt does not fit. | |||
Price per 1M tokens
Neither model changes its rate with prompt length, so these rates apply to every request.
| Context band | Price | GLM 4.7 Flash | GLM 4.7 | Difference |
|---|---|---|---|---|
| Any prompt length | Input | Free | $0.60 | GLM 4.7 Flash is 100% cheaper |
| Any prompt length | Cached input | Free | $0.11 | GLM 4.7 Flash is 100% cheaper |
| Any prompt length | Output | Free | $2.20 | GLM 4.7 Flash is 100% cheaper |
Blended: GLM 4.7 Flash Free, GLM 4.7 $1.00 per 1M โ one rate at 3:1 input to output, for comparing two models at a glance.
Both priced by Z.ai. Standard tier, pay-as-you-go.
Benchmarks
They share 17 figures: GLM 4.7 Flash leads on 2, GLM 4.7 on 15. The widest gap is 105 points, on Svg Elo.
Overall intelligence
Intelligence Index โ Overall intelligence across reasoning, knowledge & math evals
GLM 4.7 ahead by 7.3
Graduate-level science
GPQA Diamond โ Graduate-level questions in biology, chemistry & physics
GLM 4.7 ahead by 27.8
Expert-exam performance
Humanity's Last Exam โ Expert-level questions across many academic domains
GLM 4.7 ahead by 19.8
Instruction following
IFBench โ Precise following of detailed instructions
GLM 4.7 ahead by 7.1
Support-agent performance
ฯยฒ-Bench Telecom โ Tool-using agent tasks in a telecom support setting
GLM 4.7 Flash ahead by 2.9
Long-context reasoning
AA-LCR โ Long-context reasoning across large inputs
GLM 4.7 ahead by 29.3
Command-line work
Terminal-Bench Hard โ Complex command-line and terminal workflows
GLM 4.7 ahead by 9.8
Physics research reasoning
CritPt โ Unpublished physics research reasoning problems
GLM 4.7 ahead by 1.4
Factual accuracy
Omniscience Accuracy โ Breadth of factual knowledge across domains
GLM 4.7 ahead by 13.1
Answer reliability
Omniscience Non-Hallucination โ How reliably the model avoids fabricated answers
GLM 4.7 ahead by 1
3d Elo
Design Arena head-to-head rating in the 3d category
GLM 4.7 ahead by 70
Codecategories Elo
Design Arena head-to-head rating in the codecategories category
GLM 4.7 ahead by 39
Dataviz Elo
Design Arena head-to-head rating in the dataviz category
GLM 4.7 ahead by 72
Gamedev Elo
Design Arena head-to-head rating in the gamedev category
GLM 4.7 ahead by 54
Svg Elo
Design Arena head-to-head rating in the svg category
GLM 4.7 ahead by 105
Uicomponent Elo
Design Arena head-to-head rating in the uicomponent category
GLM 4.7 Flash ahead by 8
Website Elo
Design Arena head-to-head rating in the website category
GLM 4.7 ahead by 32
Each figure is that model's best published run, with the effort level named beside it โ where these scores come from.
Specifications
GLM 4.7 holds 4,800 more input tokens in one request. Only GLM 4.7 Flash lists code execution.
| Specification | GLM 4.7 Flash | GLM 4.7 |
|---|---|---|
| Context window | 200,000 tokens | 204,800 tokens |
| Max output | 131,072 tokens | 131,072 tokens |
| Takes in | Text | Text |
| Puts out | Text | Text |
| Released | 19 Jan 2026 | 22 Dec 2025 |
| Status | Active | Active |
| Sold by | Z.ai | Z.ai |
| Tool use | Yes | Yes |
| Structured outputs | Yes | Yes |
| Web search | Not listed | Not listed |
| Prompt caching | Yes | Yes |
| Code execution | Yes | Not listed |
| Computer use | Not listed | Not listed |
| Open weights | Yes | Yes |
| Reasoning | Yes | Yes |
FAQs
- Is GLM 4.7 Flash cheaper than GLM 4.7?
- Yes. A 1,000-token prompt with a 500-token reply costs $0.0000 on GLM 4.7 Flash and $0.0017 on GLM 4.7, and the same model is cheaper on every workload on this page.
- How much do GLM 4.7 Flash and GLM 4.7 cost per 1M tokens?
- GLM 4.7 Flash costs Free for input and Free for output. GLM 4.7 costs $0.60 and $2.20. Standard pay-as-you-go rates, as of 19 Jan 2026.
- Which is better, GLM 4.7 Flash or GLM 4.7?
- On published benchmarks, GLM 4.7. It leads on 15 of the 17 figures both models report, including overall intelligence (Intelligence Index), where it scores 22.2 against 14.9.
- Does GLM 4.7 Flash or GLM 4.7 have a bigger context window?
- GLM 4.7, at 204,800 tokens against 200,000. That is 4,800 more input tokens in a single request.