Model comparison
Claude Haiku 4.5 vs GLM-4.6
GLM-4.6 is the stronger model overall, scoring 40.9 to 39.3 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 4 categories and GLM-4.6 in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-4.6 leads 23.1 to 15.1.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 66.6% for Claude Haiku 4.5 and 55.4% for GLM-4.6.
- GLM-4.6 is cheaper at $0.60 / $2.20 per million input/output tokens, against $1 / $5 for Claude Haiku 4.5.
- GLM-4.6 accepts more context: 205K tokens versus 200K.
- GLM-4.6 has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | GLM-4.6 | |
|---|---|---|
| Provider | Anthropic | Z.ai (Zhipu) |
| Noometry Index | 39.3 | 40.9 |
| Released | 2025-10-15 | 2025-09-30 |
| Weights | Proprietary | Open |
| Context window | 200K | 205K |
| Max output | 64K | 131K |
| Input $ / M tokens | $1 | $0.60 |
| Output $ / M tokens | $5 | $2.20 |
| Results tracked | 53 | 29 |
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: 43.9 (#71), GLM-4.6: 39.9 (#137)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| SWE-bench Verified (bash only) | 66.6% | 55.4% |
| LMArena WebDev | 1330 | 1339 |
| SciCode | 43.3% | 38.4% |
| LMArena Coding | 1454 | 1449 |
| ALE-Bench | 653.48 | 340.82 |
| SWE-bench Multilingual | 64.7% | — |
| WeirdML | 45.4% | — |
Agentic & Tool Use Claude Haiku 4.5 leads
Claude Haiku 4.5: 33.5 (#50), GLM-4.6: 32.1 (#60)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| Terminal-Bench | 35.5% | 24.5% |
| Berkeley Function Calling Leaderboard | 68.7% | 72.4% |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning GLM-4.6 leads
Claude Haiku 4.5: 15.1 (#310), GLM-4.6: 23.1 (#168)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| CritPt | 0% | 1.1% |
| LMArena Hard Prompts | 1420 | 1440 |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | — | 47.4% |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| Chess Puzzles | 8% | — |
| DTBench | 73.6% | — |
| LMCA | 30.9% | — |
| Epoch Capabilities Index | 142.42 | — |
| ForecastBench | 61.4 | — |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.0 (#69), GLM-4.6: 36.4 (#115)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| LMArena Math | 1395 | 1431 |
| FrontierMath (Feb 2025 set) | 5.9% | 3.8% |
| FrontierMath Tier 4 (v1) | 2.1% | 2.1% |
| OTIS Mock AIME 2024-2025 | 66.7% | — |
| Omni-MATH | 56.1% | — |
| MATH Level 5 | 96.4% | — |
Knowledge GLM-4.6 leads
Claude Haiku 4.5: 37.6 (#146), GLM-4.6: 39.6 (#118)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| Vectara Hallucination Rate | 9.8% | 9.5% |
| LMArena Expert | 1443 | 1432 |
| GPQA Diamond | 71.2% | — |
| SimpleQA Verified | 13.2% | — |
| MMLU-Pro | 77.7% | — |
| GPQA (HELM) | 60.5% | — |
Multimodal Not comparable
Claude Haiku 4.5: 27.0 (#111), GLM-4.6: —
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual GLM-4.6 leads
Claude Haiku 4.5: 50.1 (#124), GLM-4.6: 53.6 (#62)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| LMArena Non-English | 1376 | 1426 |
| LMArena Chinese | 1415 | 1500 |
| LMArena French | 1408 | 1459 |
| LMArena German | 1377 | 1447 |
| LMArena Japanese | 1339 | 1394 |
| LMArena Korean | 1346 | 1401 |
| LMArena Russian | 1380 | 1418 |
| LMArena Spanish | 1420 | 1436 |
Instruction Following GLM-4.6 leads
Claude Haiku 4.5: 71.6 (#144), GLM-4.6: 74.5 (#93)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| LMArena Instruction Following | 1414 | 1410 |
| IFEval | 80.1% | — |
Long Context Too close to call
Claude Haiku 4.5: 43.6 (#86), GLM-4.6: 43.4 (#91)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| LMArena Longer Query | 1427 | 1423 |
Writing & Preference GLM-4.6 leads
Claude Haiku 4.5: 58.1 (#119), GLM-4.6: 61.3 (#86)
| Benchmark | Claude Haiku 4.5 | GLM-4.6 |
|---|---|---|
| LMArena Text | 1396 | 1440 |
| LMArena Creative Writing | 1371 | 1411 |
| LMArena Multi-Turn | 1409 | 1427 |
| EQ-Bench Creative Writing | — | 1411 |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than GLM-4.6?
GLM-4.6 is the stronger model overall, scoring 40.9 to 39.3 on the Noometry Index.
Which is cheaper, Claude Haiku 4.5 or GLM-4.6?
GLM-4.6 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; Claude Haiku 4.5 lists at $1 and $5.
Is Claude Haiku 4.5 or GLM-4.6 better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 43.9 versus 39.9 in the Noometry coding category.
Which has the bigger context window?
GLM-4.6 does, with 205K tokens against 200K.
How many benchmarks do Claude Haiku 4.5 and GLM-4.6 share?
27 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and GLM-4.6 has 29.