BlogLLM Benchmark Scores as Published, Side by Side
GLM-5.3-Flash: Benchmark Scores as Published
Every result the catalog holds for GLM-5.3-Flash: 44 tests in 14 areas, from 5 pages. 6 printed by Z.ai itself, the rest by the company named beside each number. Each number is as printed; nothing is ranked or rescored.
Developer Z.aiReleased 2026-08-26Context window 1MMax output 128KInput Text, image, videoOutput TextWeights Open (MIT)Data version 2026103001, as of 2026-10-30
Knowledge Work
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| AutomationBench | 48.8% | Z.ai, Z.ai GLM-5.3-Flash model cardchecked 2026-10-15; printed on 1 more page | Effort: max |
| GDPval-AA v2 | 1773 | Z.ai, Z.ai GLM-5.3-Flash model cardchecked 2026-10-15; printed on 1 more page | Effort: max |
Agentic Coding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| SWE-bench Verified | 79.1% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| SWE-Bench Pro | 59.7% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Terminal-Bench 2.1 | 84.3% | Z.ai, Z.ai GLM-5.3-Flash model cardchecked 2026-10-15; printed on 1 more page | Effort: max; Harness: Claude Code 2.1.207 |
| DeepSWE v1.1 | 63.4% | Z.ai, Z.ai GLM-5.3-Flash model cardchecked 2026-10-15; printed on 1 more page | Effort: max; Harness: mini-swe-agent |
| SWE-bench Multilingual | 78.3% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Terminal-Bench 4.0 | 15.15% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Science and Math
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPQA Diamond | 91.2% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| SciCode | 50.05% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| MathArena Apex Shortlist | 77.1% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Long Context
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| AA-LCR | 79.5% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| GraphWalks (variant not stated) | 88.4% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Computer Use
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| OSWorld-Verified | 14.78% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Multimodal Understanding
Cybersecurity
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Artificial Analysis Cyber Index (successes) | 50 | Mistral AI, Mistral Large 4 (preview)checked 2026-10-15 | - |
| CyberGym-E2E | 74% | Mistral AI, Mistral Large 4 (preview)checked 2026-10-15 | - |
General Knowledge and Reasoning
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Humanity's Last Exam (no tools) | 36.47% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Humanity's Last Exam (with tools) | 55.3% | Z.ai, Z.ai GLM-5.3-Flash model cardchecked 2026-10-15 | Effort: max; Tools: tools |
| AA-Omniscience Accuracy | 34.2% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| AA-Omniscience Index | 14.3 | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| AA-Omniscience Hallucination Ratelower is better | 30.3% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Humanity's Last Exam (full set, tools not stated) | 35.1% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Code Generation
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| LiveCodeBench v6 | 80.6% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Tool Use and Workflows
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| τ³-bench Banking | 34.8% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| τ²-bench Telecom | 98.5% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Agents' Last Exam (pass rate) | 26.3% | Z.ai, Z.ai GLM-5.3-Flash model cardchecked 2026-10-15 | Effort: max |
| τ²-bench Retail | 78.2% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| τ²-bench Airline | 81.8% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Web Search and Research
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| BrowseComp | 86.6% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| WebVoyager | 72.64% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Instruction Following
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| IFBench | 57.7 | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Multilingual
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| MMLU-ProX | 83% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| WMT24++ (XCOMET-XXL) | 86.1 | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Writing and Conversation
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| LMArena v3 proxy | 61 | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
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 | $0.15 | Z.ai, Z.AI API pricingchecked 2026-10-15 | - |
| Cached input | $0.03 | Z.ai, Z.AI API pricingchecked 2026-10-15 | - |
| Output | $0.5 | Z.ai, Z.AI API pricingchecked 2026-10-15 | - |
| Cost per taskVals AI, USD, lower is better | $2.35 | 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.