Model Details
Where is GLM 5.1 a perfect fit?
- Long-running autonomous coding agents
- Repository-scale software engineering
- Complex debugging and refactoring
- Terminal and CLI-based development
- Performance optimization and benchmarking
- Multi-step tool-driven workflows
- Large codebase analysis
- Autonomous research and technical investigation
- Front-end and artifact generation
- Agent systems requiring sustained execution
GLM-5.1's defining feature is sustained autonomous execution rather than simply stronger single-turn reasoning. Z.ai specifically designed it to maintain a goal across extended coding sessions, repeatedly test and evaluate its work, and optimize toward a production-ready result.
Quick Model Estimate
Your GLM 5.1 Cost Estimate
💰 Total Cost
—
for 1000 input + 1000 output tokens
Cost Breakdown
Prices are watched for changes and checked against the provider's own page.
Pricing
Benchmarks
Scores from standardized evaluations by Artificial Analysis and Design Arena. Higher is better — the indices summarize overall ability, while the detailed scores break down performance on individual benchmarks.
Artificial Analysis
Higher is better · benchmarked by Artificial AnalysisGraduate-level questions in biology, chemistry & physics
Expert-level questions across many academic domains
Precise following of detailed instructions
Tool-using agent tasks in a telecom support setting
Unpublished physics research reasoning problems
Long-context reasoning across large inputs
Research-level scientific coding tasks
Complex command-line and terminal workflows
Economically valuable, real-world knowledge work
Breadth of factual knowledge across domains
How reliably the model avoids fabricated answers
| Variant | GPQA Diamond | Humanity's Last Exam | IFBench | τ²-Bench Telecom | AA-LCR | SciCode | Terminal-Bench Hard | CritPt | GDPval | Omniscience Accuracy | Omniscience Non-Hallucination |
|---|---|---|---|---|---|---|---|---|---|---|---|
| GLM-5.1 (Reasoning) | 86.8% | 30.1% | 76.3% | 97.7% | 73.7% | 44.8% | 43.2% | 4.6% | 30.8% | 23.7% | 70.1% |
| GLM-5.1 (Non-reasoning) | 83.9% | 27.9% | 52% | 97.1% | 53.3% | — | 35.6% | 0% | — | 25.2% | 36.4% |
Design Arena
Elo rating by arena & category| Variant | Arena | Category | Elo | Win % | Percentile | Avg time (ms) |
|---|---|---|---|---|---|---|
| glm-5.1 | agents | agenticgamedev | 1,128 | 49.5% | 29 | — |
| glm-5.1 | agents | agentichtmlslides | 1,205 | 52% | 44 | — |
| glm-5.1 | agents | agenticslides | 1,245 | 54.4% | 78 | — |
| glm-5.1 | agents | agenticslides(html) | 1,204 | 51.8% | 44 | — |
| glm-5.1 | agents | agenticslides(python-pptx) | 1,240 | 53.3% | 67 | — |
| glm-5.1 | agents | androidnative | 1,183 | 52.5% | 58 | — |
| glm-5.1 | agents | fullstack | 1,167 | 53.2% | 45 | — |
| glm-5.1 | agents | godotgamedev | 1,053 | 37.8% | 6 | — |
| glm-5.1 | agents | htmlslides | 1,066 | 48.8% | 36 | 376,473 |
| glm-5.1 | agents | mobileapps | 1,160 | 50.2% | 42 | — |
| glm-5.1 | agents | pptxslides | 1,241 | 53.5% | 63 | — |
| glm-5.1 | agents | python-pptxslides | 1,144 | 50% | 43 | 582,107 |
| glm-5.1 | agents | spreadsheet | 1,315 | 72.4% | 88 | — |
| glm-5.1 | agents | webapps | 1,175 | 50.5% | 35 | — |
| glm-5.1 | models | 3d | 1,336 | 62.6% | 91 | 292,283 |
| coconut | models | 3d | 1,247 | 53.2% | 76 | 388,502 |
| coconut | models | asciiart | 1,139 | 44.6% | 37 | 329,221 |
| coconut | models | codecategories | 1,267 | 53% | 78 | 363,268 |
| glm-5.1 | models | dataviz | 1,366 | 67% | 98 | 259,663 |
| coconut | models | dataviz | 1,260 | 53.5% | 77 | 287,014 |
| coconut | models | gamedev | 1,263 | 54.8% | 79 | 446,216 |
| coconut | models | svg | 1,216 | 55.3% | 78 | 190,319 |
| glm-5.1 | models | uicomponent | 1,292 | 54.9% | 85 | 256,840 |
| coconut | models | uicomponent | 1,265 | 53.3% | 76 | 263,888 |
| glm-5.1 | models | website | 1,290 | 55.1% | 87 | 302,826 |
| coconut | models | website | 1,265 | 52.1% | 78 | 345,161 |
Compare GLM 5.1 with
Two models side by side: worked costs, every rate, and every benchmark figure both of them report.
FAQs about GLM 5.1
How much does GLM 5.1 cost per 1M tokens?
GLM 5.1 costs $1.40 per million input tokens, and $4.40 per million output tokens.
What does a typical workload cost with GLM 5.1?
1,000 requests of 2,000 input and 500 output tokens each — 2,000,000 input and 500,000 output tokens in total — costs $5.00 with GLM 5.1 at its lowest rates. Use the calculator on this page for your own volumes.
Where can I use GLM 5.1?
GLM 5.1 is available through Z.ai, from $1.40 per million input tokens.
What is GLM 5.1's context window?
GLM 5.1 has a context window of 200,000 tokens, and returns up to 128,000 tokens in a single response.
What input types does GLM 5.1 support?
GLM 5.1 accepts text input, and returns text.