GLM-5.3-Flash
| Released | 2026-08-26 |
|---|---|
| Price per 1M | $0.30 in · $1.05 out |
| Overall score | 58.3 · rank 50 |
| Context window | 1,000,000 tokens |
| Provider | Zhipu AI |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 58.3 | 50 | 45 |
| Overall: Coding | 58.4 | 47 | 36 |
| Overall: Reasoning | 57.6 | 60 | 14 |
| Overall: Agentic work | 59.2 | 40 | 33 |
| Overall: Frontend | 56.1 | 55 | 16 |
| Coding | 58.9 | 35 | 11 |
| Agents | 59.3 | 38 | 14 |
| Reasoning | 59.5 | 51 | 8 |
| Math | 56.3 | 60 | 4 |
| Knowledge | 55.8 | 53 | 2 |
| Frontend | 54.3 | 58 | 3 |
| General | 59.0 | 35 | 1 |
| Multimodal | 58.5 | 28 | 2 |
| Value vs price | +12.7 | 17 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 59.2 ±2.1 | 13 |
| High | 57.1 ±4.0 | 3 |
| Low | 54.7 ±4.0 | 3 |
| Max | 57.7 ±1.0 | 32 |
| Non-reasoning | 61.2 ±5.1 | 1 |
Evidence
Coding 58.9 ±1.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CursorBench | 36.8% | Epoch AI | 2026-11-11 |
| FrontierSWE | 18.1% | Epoch AI | 2026-11-11 |
| DeepSWE | 63.4% | Epoch AI | 2026-11-11 |
| FrontierCode | 31.8% | Epoch AI | 2026-11-11 |
| Code Migration v1.1 | 51.6% | Vals AI | 2026-11-09 |
| LiveBench 2.0 Coding | 70.9% | LiveBench | 2026-11-06 |
| IOI | 52.5% | Vals AI | 2026-10-29 |
| SWE-bench Verified (Vals run) | 92.0% | Vals AI | 2026-09-01 |
| SciCode | 51.6% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 80.5% | Vals AI | 2026-09-01 |
| ProgramBench | 0.0% | Vals AI | 2026-11-09 |
Agents 59.3 ±1.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 4 | 35.5% | Terminal-Bench | 2026-08-26 |
| SRE Bench | 0.8% | Vals AI | 2026-11-05 |
| Terminal-Bench-Science | 2.9% | Vals AI | 2026-11-09 |
| Terminal-Bench 4.0 (Vals run) | 37.4% | Vals AI | 2026-11-09 |
| APEX-Agents | 52.8% | Epoch AI | 2026-11-11 |
| LiveBench 2.0 Agentic | 20.7% | LiveBench | 2026-11-06 |
| LiveBench 2.0 Browser Use | 59.8% | LiveBench | 2026-11-06 |
| Terminal-Bench 2.1 (Vals run) | 62.9% | Vals AI | 2026-10-01 |
| EMB v1.1 (Excel financial modeling) | 66.3% | Vals AI | 2026-11-09 |
| LiveBench 2.0 Research | 73.9% | LiveBench | 2026-11-06 |
| Tax Agent Bench | 31.6% | Vals AI | 2026-11-09 |
| Finance Agent v2.1 | 60.6% | Vals AI | 2026-11-09 |
| SkillsBench | 40.2% | Vals AI | 2026-10-29 |
| HLAB-Vals 1.1 (legal agent) | 10.4% | Vals AI | 2026-11-09 |
| CyberBench v1.1 | 67.9% | Vals AI | 2026-11-05 |
Reasoning 59.5 ±2.4
| Benchmark | Result | Source | Date |
|---|---|---|---|
| ARC-AGI-2 | 65.8% | Epoch AI | 2026-11-11 |
| ARC-AGI-2 (ARC Prize verified) | 65.8% | ARC Prize | 2026-11-11 |
| CritPt | 15.4% | Epoch AI | 2026-11-11 |
| ARC-AGI-1 (ARC Prize verified) | 91.0% | ARC Prize | 2026-11-11 |
| ARC-AGI-1 | 91.0% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 90.2% | Epoch AI | 2026-08-26 |
| GPQA Diamond (Vals run) | 86.4% | Vals AI | 2026-09-01 |
| Chess Puzzles | 21.0% | Epoch AI | 2026-10-13 |
| LiveBench 2.0 Reasoning | 77.6% | LiveBench | 2026-11-06 |
| Mystery Game Puzzles | 8.0% | Epoch AI | 2026-08-28 |
| LiveBench 2.0 Data Analysis | 76.4% | LiveBench | 2026-11-06 |
Math 56.3 ±1.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| FrontierMath Tiers 1-3 | 55.8% | Epoch AI | 2026-08-27 |
| FrontierMath Tier 4 | 17.1% | Epoch AI | 2026-08-27 |
| OTIS Mock AIME 2024-2025 | 93.9% | Epoch AI | 2026-08-26 |
| ProofBench v1.1 | 21.0% | Vals AI | 2026-11-09 |
| ProofBench | 21.0% | Epoch AI | 2026-11-11 |
Knowledge 55.8 ±3.9
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMLU-Pro (Vals run) | 86.1% | Vals AI | 2026-09-01 |
| Legal Research Bench | 45.2% | Vals AI | 2026-11-09 |
Frontend 54.3 ±5.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LiveBench 2.0 Design | 73.7% | LiveBench | 2026-11-06 |
| Vibe Code Bench v1.1 | 30.8% | Vals AI | 2026-11-09 |
| LMArena WebDev | 1,609 | LMArena | 2026-11-09 |
General 59.0 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,474 | LMArena | 2026-11-09 |
Multimodal 58.5 ±2.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMMU-Pro (Vals run) | 86.0% | Vals AI | 2026-09-01 |
| LMArena Vision | 1,272 | LMArena | 2026-11-09 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| ARC-AGI-1 | 71.8% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 | 47.0% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 71.8% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-1 (ARC Prize verified) | 47.0% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 | 50.1% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 | 27.9% | Epoch AI | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 50.1% | ARC Prize | A stronger configuration result was used |
| ARC-AGI-2 (ARC Prize verified) | 27.9% | ARC Prize | A stronger configuration result was used |
| CursorBench | 31.1% | Epoch AI | A stronger configuration result was used |
| CursorBench | 26.9% | Epoch AI | A stronger configuration result was used |
Names matched to this model
glm-5-3-flash, glm-5-3-flashx