Model comparison
Claude 3.5 Haiku vs GLM-5.3-Flash
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 29.2 on the Noometry Index.
Last verified . 23 shared benchmarks.
Summary
- They share 23 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 0 categories and GLM-5.3-Flash in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5.3-Flash leads 58.4 to 18.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 93.9% for GLM-5.3-Flash.
- GLM-5.3-Flash has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Haiku | GLM-5.3-Flash | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 29.2 | 51.8 |
| Released | 2024-10-22 | 2026-08-20 |
| Weights | Proprietary | Open |
| Context window | — | 1M |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.50 |
| Results tracked | 49 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3-Flash leads
Claude 3.5 Haiku: 32.9 (#265), GLM-5.3-Flash: 53.1 (#31)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| SciCode | 27.4% | 51.6% |
| LMArena Coding | 1286 | 1508 |
| DeepSWE | — | 63.4% |
| FrontierCode | — | 31.8% |
| Aider Polyglot | 28% | — |
| CursorBench | — | 36.8% |
| LMArena WebDev | — | 1609 |
| FrontierSWE | — | 18.1% |
| WeirdML | 30.7% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 303.55 |
Agentic & Tool Use GLM-5.3-Flash leads
Claude 3.5 Haiku: 28.0 (#95), GLM-5.3-Flash: 34.2 (#47)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| APEX-Agents | — | 52.8% |
| BALROG | 19.3% | — |
| GDP.pdf | — | 14% |
Reasoning GLM-5.3-Flash leads
Claude 3.5 Haiku: 17.7 (#290), GLM-5.3-Flash: 48.0 (#42)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| CritPt | 0% | 15.4% |
| LMArena Hard Prompts | 1251 | 1491 |
| Epoch Capabilities Index | 127.15 | 151.88 |
| ARC-AGI-2 | — | 65.8% |
| ARC-AGI-1 | — | 91% |
| Chess Puzzles | — | 14% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 8% |
| DTBench | 56.7% | — |
| LiveBench Data Analysis | 48.5% | — |
| Surface Evolver Bench | — | 52.5% |
| Bench to the Future 3 | — | 0.15 |
| LiveBench | 43.5% | — |
Math GLM-5.3-Flash leads
Claude 3.5 Haiku: 14.7 (#300), GLM-5.3-Flash: 53.3 (#47)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 93.9% |
| LMArena Math | 1244 | 1500 |
| FrontierMath (Tiers 1-3) | — | 55.8% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 21% |
| Omni-MATH | 22.4% | — |
| LiveBench Math | 35.5% | — |
| MATH Level 5 | 46.4% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge GLM-5.3-Flash leads
Claude 3.5 Haiku: 18.7 (#281), GLM-5.3-Flash: 58.4 (#36)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| GPQA Diamond | 38.1% | 90.2% |
| LMArena Expert | 1208 | 1513 |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| GPQA (HELM) | 36.3% | — |
| MMLU | 74.3% | — |
Multimodal GLM-5.3-Flash leads
Claude 3.5 Haiku: 26.8 (#117), GLM-5.3-Flash: 42.8 (#27)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| LMArena Vision | 1092 | 1296 |
| GeoBench | 34% | — |
Multilingual GLM-5.3-Flash leads
Claude 3.5 Haiku: 40.0 (#218), GLM-5.3-Flash: 56.0 (#25)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| LMArena Non-English | 1238 | 1462 |
| LMArena Chinese | 1229 | 1527 |
| LMArena French | 1264 | 1496 |
| LMArena German | 1237 | 1470 |
| LMArena Japanese | 1175 | 1429 |
| LMArena Korean | 1173 | 1446 |
| LMArena Russian | 1253 | 1469 |
| LMArena Spanish | 1261 | 1471 |
Instruction Following GLM-5.3-Flash leads
Claude 3.5 Haiku: 62.9 (#234), GLM-5.3-Flash: 77.5 (#20)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| LMArena Instruction Following | 1241 | 1478 |
| LiveBench Instruction Following | 61.9% | — |
| IFEval | 79.2% | — |
Long Context GLM-5.3-Flash leads
Claude 3.5 Haiku: 38.3 (#200), GLM-5.3-Flash: 45.4 (#39)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| LMArena Longer Query | 1261 | 1482 |
Writing & Preference GLM-5.3-Flash leads
Claude 3.5 Haiku: 42.7 (#234), GLM-5.3-Flash: 65.3 (#50)
| Benchmark | Claude 3.5 Haiku | GLM-5.3-Flash |
|---|---|---|
| LMArena Text | 1255 | 1471 |
| LMArena Creative Writing | 1233 | 1442 |
| LMArena Multi-Turn | 1265 | 1467 |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | 1146 | — |
| WildBench | 76% | — |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than GLM-5.3-Flash?
GLM-5.3-Flash is the stronger model overall, scoring 51.8 to 29.2 on the Noometry Index.
Is Claude 3.5 Haiku or GLM-5.3-Flash better for coding?
GLM-5.3-Flash scores higher on coding benchmarks: 53.1 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and GLM-5.3-Flash share?
23 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and GLM-5.3-Flash has 40.