Model comparison
GLM-4.6 vs GPT-5.6 Terra
GPT-5.6 Terra is the stronger model overall, scoring 59.1 to 40.9 on the Noometry Index. GLM-4.6 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 . 23 shared benchmarks.
Summary
- They share 23 benchmarks with published results for both. GLM-4.6 scores higher in 0 categories and GPT-5.6 Terra in 9 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.5 to 36.4.
- The biggest single-benchmark swing is CritPt: 1.1% for GLM-4.6 and 30% for GPT-5.6 Terra.
- GLM-4.6 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-4.6 has downloadable open weights; the other is API-only.
Side by side
| GLM-4.6 | GPT-5.6 Terra | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 40.9 | 59.1 |
| Released | 2025-09-30 | 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 | 29 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Terra leads
GLM-4.6: 39.9 (#137), GPT-5.6 Terra: 57.1 (#16)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| LMArena WebDev | 1339 | 1520 |
| SciCode | 38.4% | 55% |
| LMArena Coding | 1449 | 1484 |
| ALE-Bench | 340.82 | 1,951 |
| DeepSWE | — | 69.6% |
| FrontierCode | — | 41.3% |
| SWE-bench Verified (bash only) | 55.4% | — |
| CursorBench | — | 41.3% |
| WeirdML | — | 78.3% |
Agentic & Tool Use GPT-5.6 Terra leads
GLM-4.6: 32.1 (#60), GPT-5.6 Terra: 40.2 (#23)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| Terminal-Bench | 24.5% | — |
| APEX-Agents | — | 58.2% |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| BALROG | — | 53.2% |
| GDP.pdf | — | 24.7% |
| Vending-Bench 2 | — | 7,343 |
Reasoning GPT-5.6 Terra leads
GLM-4.6: 23.1 (#168), GPT-5.6 Terra: 60.6 (#19)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| Kagi LLM Benchmark | 47.4% | 51.3% |
| CritPt | 1.1% | 30% |
| LMArena Hard Prompts | 1440 | 1467 |
| ARC-AGI-2 | — | 83.9% |
| SimpleBench | — | 48.9% |
| NYT Connections (extended) | — | 78.4% |
| ARC-AGI-1 | — | 96.5% |
| Chess Puzzles | — | 54% |
| Mystery Game Puzzles | — | 35% |
| DTBench | — | 93.3% |
| LMCA | — | 55% |
| Surface Evolver Bench | — | 83.8% |
| Epoch Capabilities Index | — | 159.31 |
Math GPT-5.6 Terra leads
GLM-4.6: 36.4 (#115), GPT-5.6 Terra: 81.5 (#10)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| LMArena Math | 1431 | 1463 |
| FrontierMath (Tiers 1-3) | — | 86% |
| FrontierMath Tier 4 | — | 70.7% |
| OTIS Mock AIME 2024-2025 | — | 99.7% |
| ProofBench | — | 74% |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5.6 Terra leads
GLM-4.6: 39.6 (#118), GPT-5.6 Terra: 61.3 (#27)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| LMArena Expert | 1432 | 1492 |
| GPQA Diamond | — | 93.3% |
| SimpleQA Verified | — | 43.2% |
| Vectara Hallucination Rate | 9.5% | — |
Multimodal Not comparable
GLM-4.6: —, GPT-5.6 Terra: 47.5 (#10)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| LMArena Vision | — | 1271 |
| Blueprint-Bench 2 | — | 30.8% |
| Furniture Assembly | — | 54.2% |
| LMArena Document | — | 1472 |
Multilingual Too close to call
GLM-4.6: 53.6 (#62), GPT-5.6 Terra: 54.5 (#42)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| LMArena Non-English | 1426 | 1439 |
| LMArena Chinese | 1500 | 1514 |
| LMArena French | 1459 | 1476 |
| LMArena German | 1447 | 1462 |
| LMArena Japanese | 1394 | 1446 |
| LMArena Korean | 1401 | 1420 |
| LMArena Russian | 1418 | 1450 |
| LMArena Spanish | 1436 | 1455 |
Instruction Following GPT-5.6 Terra leads
GLM-4.6: 74.5 (#93), GPT-5.6 Terra: 76.6 (#38)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| LMArena Instruction Following | 1410 | 1453 |
Long Context Too close to call
GLM-4.6: 43.4 (#91), GPT-5.6 Terra: 44.4 (#64)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| LMArena Longer Query | 1423 | 1452 |
Writing & Preference GPT-5.6 Terra leads
GLM-4.6: 61.3 (#86), GPT-5.6 Terra: 70.4 (#21)
| Benchmark | GLM-4.6 | GPT-5.6 Terra |
|---|---|---|
| LMArena Text | 1440 | 1446 |
| LMArena Creative Writing | 1411 | 1411 |
| EQ-Bench Creative Writing | 1411 | 1855 |
| LMArena Multi-Turn | 1427 | 1451 |
| EQ-Bench 4 | — | 1234 |
Frequently asked questions
Is GLM-4.6 better than GPT-5.6 Terra?
GPT-5.6 Terra is the stronger model overall, scoring 59.1 to 40.9 on the Noometry Index. GLM-4.6 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-4.6 or GPT-5.6 Terra?
GLM-4.6 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-4.6 or GPT-5.6 Terra better for coding?
GPT-5.6 Terra scores higher on coding benchmarks: 57.1 versus 39.9 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-4.6 and GPT-5.6 Terra share?
23 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and GPT-5.6 Terra has 52.