Model comparison
GLM-5 vs Grok 4.20 (Non-Reasoning)
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 45.9 on the Noometry Index.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. GLM-5 scores higher in 3 categories and Grok 4.20 (Non-Reasoning) in 6 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.2 to 27.6.
- The biggest single-benchmark swing is ARC-AGI-2: 4.9% for GLM-5 and 65.1% for Grok 4.20 (Non-Reasoning).
- Both cost about the same: $1 input and $3.20 output per million tokens.
- Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 205K.
- GLM-5 has downloadable open weights; the other is API-only.
Side by side
| GLM-5 | Grok 4.20 (Non-Reasoning) | |
|---|---|---|
| Provider | Z.ai (Zhipu) | xAI |
| Noometry Index | 45.9 | 48.5 |
| Released | 2026-02-11 | 2026-02-17 |
| Weights | Open | Proprietary |
| Context window | 205K | 1M |
| Max output | 131K | 30K |
| Input $ / M tokens | $1 | $1.25 |
| Output $ / M tokens | $3.20 | $2.50 |
| Results tracked | 45 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
GLM-5: 49.0 (#47), Grok 4.20 (Non-Reasoning): 41.8 (#104)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena WebDev | 1434 | 1374 |
| WeirdML | 48.2% | 52.3% |
| LMArena Coding | 1461 | 1458 |
| ALE-Bench | 765.62 | 1,150 |
| SWE-bench Verified | 72.1% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 69.7% | — |
Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads
GLM-5: 30.9 (#65), Grok 4.20 (Non-Reasoning): 34.3 (#43)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Terminal-Bench | 52.4% | 57.3% |
| τ²-bench Banking | 9.8% | 18% |
| Vending-Bench 2 | 4,432 | 4,663 |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Retail | 73.7% | — |
| τ²-bench Telecom | 86.8% | — |
| LMArena Search | — | 1189 |
Reasoning Grok 4.20 (Non-Reasoning) leads
GLM-5: 27.6 (#100), Grok 4.20 (Non-Reasoning): 52.2 (#30)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 4.9% | 65.1% |
| Kagi LLM Benchmark | 75% | 75% |
| NYT Connections (extended) | 74.8% | 85.4% |
| ARC-AGI-1 | 44.7% | 89.5% |
| Chess Puzzles | 10% | 24% |
| LMArena Hard Prompts | 1451 | 1451 |
| Epoch Capabilities Index | 145.84 | 152.01 |
| ForecastBench | 61 | 61.4 |
| SimpleBench | 53.2% | — |
| Thematic Generalization | — | 63.8% |
| DTBench | — | 90.1% |
| LMCA | — | 38.7% |
Math Grok 4.20 (Non-Reasoning) leads
GLM-5: 44.8 (#66), Grok 4.20 (Non-Reasoning): 48.0 (#54)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 92.2% |
| LMArena Math | 1440 | 1455 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 17.1% |
| MathArena Final-Answer Competitions | 65.7% | — |
| ProofBench | — | 14% |
| FrontierMath (Feb 2025 set) | 16.4% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Too close to call
GLM-5: 52.2 (#59), Grok 4.20 (Non-Reasoning): 52.8 (#54)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 87.8% | 89.3% |
| LMArena Expert | 1454 | 1440 |
| SimpleQA Verified | — | 30.2% |
| Vectara Hallucination Rate | 10.1% | — |
Multimodal Not comparable
GLM-5: —, Grok 4.20 (Non-Reasoning): 33.2 (#88)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Vision | — | 1263 |
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1416 |
Multilingual Too close to call
GLM-5: 53.9 (#57), Grok 4.20 (Non-Reasoning): 54.7 (#38)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1430 | 1440 |
| LMArena Chinese | 1511 | 1481 |
| LMArena French | 1455 | 1477 |
| LMArena German | 1445 | 1463 |
| LMArena Japanese | 1416 | 1450 |
| LMArena Korean | 1421 | 1417 |
| LMArena Russian | 1436 | 1457 |
| LMArena Spanish | 1452 | 1442 |
Instruction Following Too close to call
GLM-5: 75.4 (#64), Grok 4.20 (Non-Reasoning): 75.0 (#79)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1427 | 1420 |
Long Context Too close to call
GLM-5: 44.6 (#58), Grok 4.20 (Non-Reasoning): 45.4 (#33)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| CL-bench | 18.7% | 22.2% |
| LMArena Longer Query | 1446 | 1436 |
| CL-bench Life | — | 11.9% |
Writing & Preference Too close to call
GLM-5: 66.2 (#38), Grok 4.20 (Non-Reasoning): 65.9 (#42)
| Benchmark | GLM-5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1446 | 1451 |
| LMArena Creative Writing | 1438 | 1439 |
| EQ-Bench Creative Writing | 1601 | 1574 |
| LMArena Multi-Turn | 1456 | 1456 |
Frequently asked questions
Is GLM-5 better than Grok 4.20 (Non-Reasoning)?
Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 45.9 on the Noometry Index.
Which is cheaper, GLM-5 or Grok 4.20 (Non-Reasoning)?
GLM-5 is cheaper. It lists at $1 per million input tokens and $3.20 per million output tokens; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.
Is GLM-5 or Grok 4.20 (Non-Reasoning) better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 41.8 in the Noometry coding category.
Which has the bigger context window?
Grok 4.20 (Non-Reasoning) does, with 1M tokens against 205K.
How many benchmarks do GLM-5 and Grok 4.20 (Non-Reasoning) share?
34 benchmarks have published results for both models. GLM-5 has 45 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.