BlogLLM Benchmark Scores as Published, Side by Side
GLM-4.6: Benchmark Scores as Published
Every result the catalog holds for GLM-4.6: 16 tests in 6 areas, from 4 pages. All were printed by Z.ai itself. Each number is as printed; nothing is ranked or rescored.
Developer Z.aiReleased 2025-09-30Context window 200KMax output 128KInput TextOutput TextWeights Open (MIT)Data version 2026103001, as of 2026-10-30
Agentic Coding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| SWE-bench Verified | 68% | Z.ai, GLM-4.7 model cardchecked 2026-10-15; printed on 1 more page | - |
| SWE-bench Multilingual | 53.8% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| Terminal-Bench 2.0 | 24.5% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| Terminal-Bench Hard | 23.6% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
Science and Math
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPQA Diamond | 81% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| AIME 2025 | 93.9% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| HMMT February 2025 | 89.2% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| IMO-AnswerBench | 73.5% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| HMMT November 2025 | 87.7% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
General Knowledge and Reasoning
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Humanity's Last Exam (no tools) | 17.2% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | Tools: none |
| MMLU-Pro | 83.2% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
| Humanity's Last Exam (with tools) | 30.4% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | Tools: tools |
Code Generation
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| LiveCodeBench v6 | 82.8% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
Tool Use and Workflows
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| τ²-bench (domain not stated) | 75.2% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | - |
Web Search and Research
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| BrowseComp | 45.1% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | Tools: browsing |
| 57.5% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | Tools: browsing with context management | |
| BrowseComp-ZH | 49.5% | Z.ai, GLM-4.7 model cardchecked 2026-10-15 | Tools: browsing |
Price
API list prices in US dollars per one million tokens, as printed on the developer's pricing page on the day checked. Cost per task is measured by Vals AI.
| Price | Value | Printed by | Setup as printed |
|---|---|---|---|
| Input | $0.6 | Z.ai, Z.AI API pricingchecked 2026-10-15 | - |
| Cached input | $0.11 | Z.ai, Z.AI API pricingchecked 2026-10-15 | - |
| Output | $2.2 | Z.ai, Z.AI API pricingchecked 2026-10-15 | - |
| Cost per taskVals AI, USD, lower is better | $10.85 | Vals AI, Vibe Code Bench v1.1 leaderboardchecked 2026-10-27 | Harness: OpenHands |
More from Z.ai
Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.