Model comparison
GLM-5 vs Grok 4.20
Head-to-head evidence from 6 shared benchmark results across 4 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: GLM-5 #31 (Supported); Grok 4.20 #93 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. GLM-5 and Grok 4.20 share 6 comparable benchmark results. 3 of 8 categories are comparable. 43 results are unique to GLM-5; 13 to Grok 4.20.
Updated July 28, 2026- Shared results
- 6
- GLM-5 only
- 43
- Grok 4.20 only
- 13
- Comparable categories
- 3 / 8
Pick GLM-5 if you want the stronger benchmark profile. Grok 4.20 only becomes the better choice if coding is the priority or you need the larger 2M context window.
Confidence note. This is a partial-evidence comparison with 6 shared benchmark results across 4 evidence categories; 3 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
GLM-5 is clearly ahead on the BenchAlign aggregate, 65.24 to 53.88. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GLM-5's sharpest advantage is in agentic, where it averages 56.2 against 47.1. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 56.2% to 47.1%. Grok 4.20 does hit back in coding, so the answer changes if that is the part of the workload you care about most.
Grok 4.20 is also the more expensive model on tokens at $2.00 input / $6.00 output per 1M tokens, versus $1.00 input / $3.20 output per 1M tokens for GLM-5. Grok 4.20 is the reasoning model in the pair, while GLM-5 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Grok 4.20 gives you the larger context window at 2M, compared with 200K for GLM-5.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | GLM-5 | Δ | Grok 4.20 |
|---|---|---|---|
| Agentic | GLM-556.2 | Margin← 9.1 | Grok 4.2047.1 |
| Reasoning | GLM-560.8 | Margin← 7.5 | Grok 4.2053.3 |
| Coding | GLM-566.3 | Margin→ 0.8 | Grok 4.2067.1 |
| Knowledge | GLM-566.4 | MarginNo overlap | Grok 4.20Not measured |
| Math | GLM-556.3 | MarginNo overlap | Grok 4.20Not measured |
| Multilingual | GLM-583.1 | MarginNo overlap | Grok 4.20Not measured |
| Multimodal | GLM-5Not measured | MarginNo overlap | Grok 4.2070.1 |
| Inst. Following | GLM-592.6 | MarginNo overlap | Grok 4.20Not measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
Terminal-Bench 2.0
AgenticA 56.2%B 47.1%Winner: GLM-5Δ 9.1Terminal-Bench 2.0: GLM-5 scored 56.2%; Grok 4.20 scored 47.1%. GLM-5 wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 55.1%B 51.8%Winner: GLM-5Δ 3.3SWE-bench Pro: GLM-5 scored 55.1%; Grok 4.20 scored 51.8%. GLM-5 wins this benchmark. - Source ↗
SWE-bench Verified
CodingA 77.8%B 76.7%Winner: GLM-5Δ 1.1SWE-bench Verified: GLM-5 scored 77.8%; Grok 4.20 scored 76.7%. GLM-5 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | GLM-5 | Grok 4.20 | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | GLM-5$1 input / $3.2 output | Grok 4.20$2 input / $6 output | GLM-5 has the lower combined listed price. |
| Generation speedtokens per second | GLM-574 tok/s | Grok 4.20233 tok/s | Grok 4.20 has the higher measured throughput. |
| First-answer latencyseconds to first token | GLM-51.64 s | Grok 4.2010.33 s | GLM-5 reaches the first token sooner. |
| Context windowmaximum listed tokens | GLM-5200K | Grok 4.202M | Grok 4.20 lists the larger context window. |
Benchmark Deep Dive
AgenticGLM-5 wins14 benchmarks
| Benchmark | GLM-5 | Grok 4.20 | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 56.2% | 47.1% | GLM-5 leads |
| Claw-EvalSource | 57.7% | — | Not comparable |
| QwenClawBenchSource | 54.1% | — | Not comparable |
| τ³-bench resultsSource | 65.6% | — | Not comparable |
| DeepPlanningSource | 14.6% | — | Not comparable |
| ToolathlonSource | 38% | — | Not comparable |
| MCP AtlasSource | 31.1% | — | Not comparable |
| MCP-TasksSource | 60.8% | — | Not comparable |
| WideResearchSource | 69.8% | — | Not comparable |
| τ²-bench resultsSource | 98.2% | — | Not comparable |
| CyberGymSource | 43.2% | — | Not comparable |
| APEX-Agents-AASource | 14.5% | — | Not comparable |
| Gert LabsSource | 50.99% | 38.36% | GLM-5 leads |
| DeepSearchQASource | — | 62.8% | Not comparable |
CodingGrok 4.20 wins9 benchmarks
| Benchmark | GLM-5 | Grok 4.20 | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 77.8% | 76.7% | GLM-5 leads |
| SWE-bench Verified*Source | 72.8% | — | Not comparable |
| SWE-bench ProSource | 55.1% | 51.8% | GLM-5 leads |
| SWE MultilingualSource | 73.3% | — | Not comparable |
| SWE-RebenchSource | 62.8% | — | Not comparable |
| React Native EvalsSource | 74.8% | — | Not comparable |
| AA-SciCodeSource | 46.2% | — | Not comparable |
| LiveCodeBench ProSource | — | 74.2% | Not comparable |
| Vibe Code BenchSource | — | 4.06% | Not comparable |
ReasoningGLM-5 wins6 benchmarks
Knowledge15 benchmarks
| Benchmark | GLM-5 | Grok 4.20 | Result |
|---|---|---|---|
| GPQASource | 86% | — | Not comparable |
| GPQA-DSource | 86.0% | 88.5% | Grok 4.20 leads |
| SuperGPQASource | 66.8% | — | Not comparable |
| MMLU-ProSource | 85.7% | — | Not comparable |
| MMLU-Pro (Arcee)Source | 85.8% | — | Not comparable |
| HLESource | 50.4% | — | Not comparable |
| Artificial Analysis Intelligence IndexSource | 39.5% | — | Not comparable |
| AA-GPQA DiamondSource | 82.0% | — | Not comparable |
| AA-HLESource | 27.2% | — | Not comparable |
| AA-Omniscience IndexSource | 2.0% | — | Not comparable |
| AA-Omniscience AccuracySource | 26.9% | — | Not comparable |
| AA-Omniscience Hallucination RateSource | 34.0% | — | Not comparable |
| HLE w/o toolsSource | — | 31.6% | Not comparable |
| HealthBench HardSource | — | 20.3% | Not comparable |
| MedXpertQA (Text)Source | — | 50.2% | Not comparable |
Math8 benchmarks
| Benchmark | GLM-5 | Grok 4.20 | Result |
|---|---|---|---|
| AIME26Source | 95.8% | — | Not comparable |
| AIME25 (Arcee)Source | 93.3% | — | Not comparable |
| HMMT Feb 2025Source | 97.5% | — | Not comparable |
| HMMT Nov 2025Source | 96.9% | — | Not comparable |
| HMMT Feb 2026Source | 86.4% | — | Not comparable |
| MMAnswerBenchSource | 82.5% | — | Not comparable |
| FrontierMath v2 (Tiers 1-3)Source | 16.434% | — | Not comparable |
| FrontierMath v2 (Tier 4)Source | 2.100% | — | Not comparable |
Multilingual2 benchmarks
Multimodal6 benchmarks
Frequently Asked Questions (4)
Which is better, GLM-5 or Grok 4.20?
GLM-5 is ahead on BenchLM's BenchAlign leaderboard, 65.24 to 53.88. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 56.2% and 47.1%.
Which is better for coding, GLM-5 or Grok 4.20?
Grok 4.20 has the edge for coding in this comparison, averaging 67.1 versus 66.3. Inside this category, SWE-bench Pro is the benchmark that creates the most daylight between them.
Which is better for reasoning, GLM-5 or Grok 4.20?
GLM-5 has the edge for reasoning in this comparison, averaging 60.8 versus 53.3. Grok 4.20 stays close enough that the answer can still flip depending on your workload.
Which is better for agentic tasks, GLM-5 or Grok 4.20?
GLM-5 has the edge for agentic tasks in this comparison, averaging 56.2 versus 47.1. Inside this category, Gert Labs is the benchmark that creates the most daylight between them.
Related Comparisons
Explore More
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
One email each week. Unsubscribe anytime.