Model comparison
GLM-4.7 vs GPT-5.6 Terra
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 42.0 on the Noometry Index. GLM-4.7 costs 4.5× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. GLM-4.7 scores higher in 0 categories and GPT-5.6 Terra in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.6 to 38.6.
- The biggest single-benchmark swing is ProofBench: 6% for GLM-4.7 and 74% for GPT-5.6 Terra.
- GLM-4.7 is cheaper at $0.60 / $2.20 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
- GPT-5.6 Terra accepts more context: 1.05M tokens versus 205K.
- GLM-4.7 has downloadable open weights; the other is API-only.
Side by side
| GLM-4.7 | GPT-5.6 Terra | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 42.0 | 59.2 |
| Released | 2025-12-22 | 2026-07-09 |
| Weights | Open | Proprietary |
| Context window | 205K | 1.05M |
| Max output | 131K | 128K |
| Input $ / M tokens | $0.60 | $2 |
| Output $ / M tokens | $2.20 | $12 |
| Results tracked | 36 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Terra leads
GLM-4.7: 44.0 (#79), GPT-5.6 Terra: 57.7 (#19)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| LMArena WebDev | 1435 | 1522 |
| SciCode | 45.1% | 55% |
| LMArena Coding | 1454 | 1484 |
| ALE-Bench | 399.48 | 1,951 |
| DeepSWE | — | 69.6% |
| FrontierCode | — | 41.3% |
| CursorBench | — | 41.3% |
| WeirdML | — | 78.3% |
Agentic & Tool Use GPT-5.6 Terra leads
GLM-4.7: 26.5 (#103), GPT-5.6 Terra: 40.1 (#25)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| Vending-Bench 2 | 2,377 | 7,343 |
| Terminal-Bench | 33.4% | — |
| APEX-Agents | — | 58.2% |
| BALROG | — | 53.2% |
| GDP.pdf | — | 24.7% |
Reasoning GPT-5.6 Terra leads
GLM-4.7: 24.3 (#164), GPT-5.6 Terra: 60.7 (#21)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| SimpleBench | 47.7% | 48.9% |
| CritPt | 1.7% | 30% |
| Chess Puzzles | 6% | 54% |
| LMArena Hard Prompts | 1443 | 1468 |
| Epoch Capabilities Index | 143.51 | 159.62 |
| ARC-AGI-2 | — | 83.9% |
| Kagi LLM Benchmark | — | 51.3% |
| NYT Connections (extended) | — | 78.4% |
| ARC-AGI-1 | — | 96.5% |
| Mystery Game Puzzles | — | 35% |
| DTBench | — | 93.3% |
| LMCA | — | 55% |
| Surface Evolver Bench | — | 83.8% |
Math GPT-5.6 Terra leads
GLM-4.7: 38.6 (#135), GPT-5.6 Terra: 81.6 (#12)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 83.3% | 99.7% |
| ProofBench | 6% | 74% |
| LMArena Math | 1423 | 1466 |
| FrontierMath (Tiers 1-3) | — | 86% |
| FrontierMath Tier 4 | — | 70.7% |
| FrontierMath (Feb 2025 set) | 2.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-5.6 Terra leads
GLM-4.7: 47.0 (#80), GPT-5.6 Terra: 61.2 (#30)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| GPQA Diamond | 83.3% | 93.3% |
| SimpleQA Verified | 32.2% | 43.2% |
| LMArena Expert | 1424 | 1492 |
| Vectara Hallucination Rate | 11.7% | — |
Multimodal Not comparable
GLM-4.7: —, GPT-5.6 Terra: 47.3 (#11)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| LMArena Vision | — | 1271 |
| Blueprint-Bench 2 | — | 30.8% |
| Furniture Assembly | — | 54.2% |
| LMArena Document | — | 1472 |
Multilingual GPT-5.6 Terra leads
GLM-4.7: 52.8 (#79), GPT-5.6 Terra: 54.4 (#44)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| LMArena Non-English | 1417 | 1439 |
| LMArena Chinese | 1495 | 1513 |
| LMArena French | 1432 | 1471 |
| LMArena German | 1424 | 1460 |
| LMArena Japanese | 1439 | 1457 |
| LMArena Korean | 1399 | 1425 |
| LMArena Russian | 1423 | 1450 |
| LMArena Spanish | 1434 | 1448 |
Instruction Following GPT-5.6 Terra leads
GLM-4.7: 74.4 (#95), GPT-5.6 Terra: 76.4 (#40)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| LMArena Instruction Following | 1411 | 1454 |
Long Context GPT-5.6 Terra leads
GLM-4.7: 42.8 (#116), GPT-5.6 Terra: 44.4 (#68)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| LMArena Longer Query | 1432 | 1451 |
| CL-bench | 15.9% | — |
| CL-bench Life | 10.9% | — |
Writing & Preference GPT-5.6 Terra leads
GLM-4.7: 60.9 (#93), GPT-5.6 Terra: 70.2 (#23)
| Benchmark | GLM-4.7 | GPT-5.6 Terra |
|---|---|---|
| LMArena Text | 1435 | 1447 |
| LMArena Creative Writing | 1401 | 1410 |
| EQ-Bench Creative Writing | 1413 | 1855 |
| LMArena Multi-Turn | 1446 | 1449 |
| EQ-Bench 4 | — | 1234 |
Frequently asked questions
Is GLM-4.7 better than GPT-5.6 Terra?
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 42.0 on the Noometry Index. GLM-4.7 costs 4.5× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.
Which is cheaper, GLM-4.7 or GPT-5.6 Terra?
GLM-4.7 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; GPT-5.6 Terra lists at $2 and $12.
Is GLM-4.7 or GPT-5.6 Terra better for coding?
GPT-5.6 Terra scores higher on coding benchmarks: 57.7 versus 44.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Terra does, with 1.05M tokens against 205K.
How many benchmarks do GLM-4.7 and GPT-5.6 Terra share?
30 benchmarks have published results for both models. GLM-4.7 has 36 scored results on Noometry and GPT-5.6 Terra has 52.