Model comparison
GLM-5 vs GPT-6 Luna
GPT-6 Luna is the stronger model overall, scoring 53.3 to 46.1 on the Noometry Index.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. GLM-5 scores higher in 4 categories and GPT-6 Luna in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 46.4.
- The biggest single-benchmark swing is ARC-AGI-2: 4.9% for GLM-5 and 59.3% for GPT-6 Luna.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $1 / $3.20 for GLM-5.
- GPT-6 Luna accepts more context: 1.05M tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | GPT-6 Luna | |
|---|---|---|
| Provider | Z.ai (Zhipu) | OpenAI |
| Noometry Index | 46.1 | 53.3 |
| Released | 2026-02-11 | 2026-09-22 |
| Weights | Open | Proprietary |
| Context window | 205K | 1.05M |
| Max output | 131K | 128K |
| Input $ / M tokens | $1 | $0.10 |
| Output $ / M tokens | $3.20 | $0.50 |
| Results tracked | 45 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GLM-5: 49.0 (#52), GPT-6 Luna: 55.5 (#25)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| LMArena WebDev | 1434 | 1581 |
| LMArena Coding | 1461 | 1439 |
| ALE-Bench | 765.62 | 1,577 |
| SWE-bench Verified | 72.1% | — |
| DeepSWE | — | 66.6% |
| FrontierCode | — | 42.4% |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 69.7% | — |
| SciCode | — | 54.6% |
| WeirdML | 48.2% | — |
Agentic & Tool Use GPT-6 Luna leads
GLM-5: 31.1 (#71), GPT-6 Luna: 33.3 (#54)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| Terminal-Bench | 52.4% | — |
| APEX-Agents | — | 44.3% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 9.8% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| GDP.pdf | — | 23% |
| Vending-Bench 2 | 4,432 | — |
Reasoning GPT-6 Luna leads
GLM-5: 27.6 (#116), GPT-6 Luna: 48.2 (#41)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| ARC-AGI-2 | 4.9% | 59.3% |
| NYT Connections (extended) | 74.8% | 68.7% |
| ARC-AGI-1 | 44.7% | 86.7% |
| Chess Puzzles | 10% | 31% |
| LMArena Hard Prompts | 1452 | 1411 |
| Epoch Capabilities Index | 145.83 | 156.28 |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | 75% | — |
| CritPt | — | 19.4% |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 90.1% |
| LMCA | — | 44.5% |
| ForecastBench | 61 | — |
Math GPT-6 Luna leads
GLM-5: 46.4 (#71), GPT-6 Luna: 76.1 (#15)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 98.9% |
| LMArena Math | 1440 | 1416 |
| FrontierMath (Tiers 1-3) | — | 78.9% |
| FrontierMath Tier 4 | — | 56.1% |
| MathArena Final-Answer Competitions | 65.7% | — |
| ProofBench | — | 64% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-6 Luna leads
GLM-5: 52.3 (#64), GPT-6 Luna: 57.0 (#41)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| GPQA Diamond | 87.8% | 90.5% |
| LMArena Expert | 1454 | 1444 |
| SimpleQA Verified | — | 41.4% |
| Vectara Hallucination Rate | 10.1% | — |
Multimodal Not comparable
GLM-5: —, GPT-6 Luna: 42.4 (#30)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| LMArena Vision | — | 1217 |
| Blueprint-Bench 2 | — | 31.2% |
| Furniture Assembly | — | 44.2% |
Multilingual GLM-5 leads
GLM-5: 53.7 (#58), GPT-6 Luna: 50.5 (#117)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | 1430 | 1386 |
| LMArena Chinese | 1511 | 1433 |
| LMArena French | 1455 | 1420 |
| LMArena German | 1445 | 1369 |
| LMArena Japanese | 1416 | 1369 |
| LMArena Korean | 1423 | 1360 |
| LMArena Russian | 1436 | 1394 |
| LMArena Spanish | 1454 | 1393 |
Instruction Following Too close to call
GLM-5: 75.2 (#67), GPT-6 Luna: 74.3 (#99)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | 1428 | 1409 |
Long Context GLM-5 leads
GLM-5: 44.7 (#60), GPT-6 Luna: 43.0 (#111)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | 1446 | 1409 |
| CL-bench | 18.7% | — |
Writing & Preference GLM-5 leads
GLM-5: 66.0 (#38), GPT-6 Luna: 58.3 (#119)
| Benchmark | GLM-5 | GPT-6 Luna |
|---|---|---|
| LMArena Text | 1446 | 1391 |
| LMArena Creative Writing | 1439 | 1363 |
| LMArena Multi-Turn | 1456 | 1396 |
| EQ-Bench Creative Writing | 1601 | — |
Frequently asked questions
Is GLM-5 better than GPT-6 Luna?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 46.1 on the Noometry Index.
Which is cheaper, GLM-5 or GPT-6 Luna?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; GLM-5 lists at $1 and $3.20.
Is GLM-5 or GPT-6 Luna better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 49.0 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 205K.
How many benchmarks do GLM-5 and GPT-6 Luna share?
26 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and GPT-6 Luna has 42.