Model comparison
Claude Haiku 4.5 vs GLM-4.5-Air
Claude Haiku 4.5 and GLM-4.5-Air score almost the same on the Noometry Index (39.5 vs 38.9), so choose on price, context window or the category you care about most.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 7 categories and GLM-4.5-Air in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude Haiku 4.5 leads 44.0 to 33.3.
- The biggest single-benchmark swing is Omni-MATH: 56.1% for Claude Haiku 4.5 and 39.1% for GLM-4.5-Air.
- GLM-4.5-Air is cheaper at $0.20 / $1.10 per million input/output tokens, against $1 / $5 for Claude Haiku 4.5.
- Claude Haiku 4.5 accepts more context: 200K tokens versus 131K.
- GLM-4.5-Air has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | GLM-4.5-Air | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 39.5 | 38.9 |
| Released | 2025-10-15 | 2025-07-20 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 64K | 98K |
| Input $ / M tokens | $1 | $0.20 |
| Output $ / M tokens | $5 | $1.10 |
| Results tracked | 53 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Haiku 4.5 leads
Claude Haiku 4.5: 44.0 (#78), GLM-4.5-Air: 33.3 (#259)
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Coding | 1453 | 1397 |
| SWE-bench Verified (bash only) | 66.6% | — |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| GSO | — | 2.9% |
| WeirdML | 45.4% | — |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.6 (#52), GLM-4.5-Air: —
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning GLM-4.5-Air leads
Claude Haiku 4.5: 15.1 (#320), GLM-4.5-Air: 24.1 (#166)
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Hard Prompts | 1420 | 1379 |
| ForecastBench | 61.4 | 59.2 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | — | 43% |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| CritPt | 0% | — |
| Chess Puzzles | 8% | — |
| DTBench | 73.6% | — |
| LMCA | 30.9% | — |
| Epoch Capabilities Index | 142.41 | — |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 44.9 (#78), GLM-4.5-Air: 36.2 (#170)
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| Omni-MATH | 56.1% | 39.1% |
| LMArena Math | 1396 | 1396 |
| OTIS Mock AIME 2024-2025 | 66.7% | — |
| MATH Level 5 | 96.4% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Claude Haiku 4.5 leads
Claude Haiku 4.5: 37.7 (#153), GLM-4.5-Air: 35.0 (#191)
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| MMLU-Pro | 77.7% | 76.2% |
| Vectara Hallucination Rate | 9.8% | 9.3% |
| GPQA (HELM) | 60.5% | 59.4% |
| LMArena Expert | 1442 | 1370 |
| GPQA Diamond | 71.2% | — |
| Humanity's Last Exam | — | 8.1% |
| SimpleQA Verified | 13.2% | — |
Multimodal Not comparable
Claude Haiku 4.5: 26.8 (#118), GLM-4.5-Air: —
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Too close to call
Claude Haiku 4.5: 49.9 (#129), GLM-4.5-Air: 49.1 (#135)
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Non-English | 1377 | 1366 |
| LMArena Chinese | 1417 | 1426 |
| LMArena French | 1408 | 1399 |
| LMArena German | 1375 | 1377 |
| LMArena Japanese | 1339 | 1348 |
| LMArena Korean | 1347 | 1308 |
| LMArena Russian | 1381 | 1373 |
| LMArena Spanish | 1420 | 1386 |
Instruction Following Claude Haiku 4.5 leads
Claude Haiku 4.5: 71.4 (#149), GLM-4.5-Air: 69.6 (#171)
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| IFEval | 80.1% | 81.2% |
| LMArena Instruction Following | 1414 | 1354 |
Long Context Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.6 (#92), GLM-4.5-Air: 41.6 (#135)
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Longer Query | 1427 | 1366 |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 57.9 (#123), GLM-4.5-Air: 55.9 (#139)
| Benchmark | Claude Haiku 4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Text | 1396 | 1384 |
| LMArena Creative Writing | 1372 | 1343 |
| WildBench | 83.9% | 78.9% |
| LMArena Multi-Turn | 1409 | 1371 |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than GLM-4.5-Air?
Claude Haiku 4.5 and GLM-4.5-Air score almost the same on the Noometry Index (39.5 vs 38.9), so choose on price, context window or the category you care about most.
Which is cheaper, Claude Haiku 4.5 or GLM-4.5-Air?
GLM-4.5-Air is cheaper. It lists at $0.20 per million input tokens and $1.10 per million output tokens; Claude Haiku 4.5 lists at $1 and $5.
Is Claude Haiku 4.5 or GLM-4.5-Air better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 44.0 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
Claude Haiku 4.5 does, with 200K tokens against 131K.
How many benchmarks do Claude Haiku 4.5 and GLM-4.5-Air share?
24 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and GLM-4.5-Air has 27.