Model comparison
GLM-4.6 vs GLM-5
GLM-5 is the stronger model overall, scoring 46.1 to 41.4 on the Noometry Index. GLM-4.6 costs 1.6× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. GLM-4.6 scores higher in 1 category and GLM-5 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5 leads 52.3 to 40.2.
- The biggest single-benchmark swing is Terminal-Bench: 24.5% for GLM-4.6 and 52.4% for GLM-5.
- GLM-4.6 is cheaper at $0.60 / $2.20 per million input/output tokens, against $1 / $3.20 for GLM-5.
Side by side
| GLM-4.6 | GLM-5 | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Z.ai (Zhipu) |
| Noometry Index | 41.4 | 46.1 |
| Released | 2025-09-30 | 2026-02-11 |
| Weights | Open | Open |
| Context window | 205K | 205K |
| Max output | 131K | 131K |
| Input $ / M tokens | $0.60 | $1 |
| Output $ / M tokens | $2.20 | $3.20 |
| Results tracked | 29 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-4.6: 40.1 (#148), GLM-5: 49.0 (#52)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| SWE-bench Verified (bash only) | 55.4% | 72.8% |
| LMArena WebDev | 1340 | 1434 |
| LMArena Coding | 1449 | 1461 |
| ALE-Bench | 340.82 | 765.62 |
| SWE-bench Verified | — | 72.1% |
| SWE-bench Multilingual | — | 69.7% |
| SciCode | 38.4% | — |
| WeirdML | — | 48.2% |
Agentic & Tool Use GLM-4.6 leads
GLM-4.6: 32.3 (#66), GLM-5: 31.1 (#71)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| Terminal-Bench | 24.5% | 52.4% |
| Berkeley Function Calling Leaderboard | 72.4% | — |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 73.7% |
| τ²-bench Telecom | — | 86.8% |
| Vending-Bench 2 | — | 4,432 |
Reasoning GLM-5 leads
GLM-4.6: 23.7 (#172), GLM-5: 27.6 (#116)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| Kagi LLM Benchmark | 47.4% | 75% |
| LMArena Hard Prompts | 1440 | 1452 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | — | 53.2% |
| NYT Connections (extended) | — | 74.8% |
| ARC-AGI-1 | — | 44.7% |
| CritPt | 1.1% | — |
| Chess Puzzles | — | 10% |
| Epoch Capabilities Index | — | 145.83 |
| ForecastBench | — | 61 |
Math GLM-5 leads
GLM-4.6: 39.1 (#111), GLM-5: 46.4 (#71)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| LMArena Math | 1432 | 1440 |
| FrontierMath (Feb 2025 set) | 3.8% | 16.4% |
| FrontierMath Tier 4 (v1) | 2.1% | 2.1% |
| MathArena Final-Answer Competitions | — | 65.7% |
| OTIS Mock AIME 2024-2025 | — | 80% |
Knowledge GLM-5 leads
GLM-4.6: 40.2 (#124), GLM-5: 52.3 (#64)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| Vectara Hallucination Rate | 9.5% | 10.1% |
| LMArena Expert | 1431 | 1454 |
| GPQA Diamond | — | 87.8% |
Multilingual Too close to call
GLM-4.6: 53.5 (#66), GLM-5: 53.7 (#58)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| LMArena Non-English | 1426 | 1430 |
| LMArena Chinese | 1499 | 1511 |
| LMArena French | 1459 | 1455 |
| LMArena German | 1447 | 1445 |
| LMArena Japanese | 1393 | 1416 |
| LMArena Korean | 1400 | 1423 |
| LMArena Russian | 1419 | 1436 |
| LMArena Spanish | 1436 | 1454 |
Instruction Following Too close to call
GLM-4.6: 74.3 (#98), GLM-5: 75.2 (#67)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| LMArena Instruction Following | 1410 | 1428 |
Long Context GLM-5 leads
GLM-4.6: 43.4 (#94), GLM-5: 44.7 (#60)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| LMArena Longer Query | 1422 | 1446 |
| CL-bench | — | 18.7% |
Writing & Preference GLM-5 leads
GLM-4.6: 61.1 (#90), GLM-5: 66.0 (#38)
| Benchmark | GLM-4.6 | GLM-5 |
|---|---|---|
| LMArena Text | 1440 | 1446 |
| LMArena Creative Writing | 1411 | 1439 |
| EQ-Bench Creative Writing | 1411 | 1601 |
| LMArena Multi-Turn | 1427 | 1456 |
Frequently asked questions
Is GLM-4.6 better than GLM-5?
GLM-5 is the stronger model overall, scoring 46.1 to 41.4 on the Noometry Index. GLM-4.6 costs 1.6× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Which is cheaper, GLM-4.6 or GLM-5?
GLM-4.6 is cheaper. It lists at $0.60 per million input tokens and $2.20 per million output tokens; GLM-5 lists at $1 and $3.20.
Is GLM-4.6 or GLM-5 better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
Both accept 205K tokens.
How many benchmarks do GLM-4.6 and GLM-5 share?
26 benchmarks have published results for both models. GLM-4.6 has 29 scored results on Noometry and GLM-5 has 45.