Model comparison
GLM-5 vs Qwen3.6-27B
Head-to-head evidence from 27 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: GLM-5 #28 (Supported); Qwen3.6-27B #93 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. GLM-5 and Qwen3.6-27B share 27 comparable benchmark results. 4 of 8 categories are comparable. 22 results are unique to GLM-5; 27 to Qwen3.6-27B.
Updated July 21, 2026- Shared results
- 27
- GLM-5 only
- 22
- Qwen3.6-27B only
- 27
- Comparable categories
- 4 / 8
Pick GLM-5 if you want the stronger benchmark profile. Qwen3.6-27B only becomes the better choice if mathematics is the priority or you want the cheaper token bill.
Confidence note. This is a partial-evidence comparison with 27 shared benchmark results across 6 evidence categories; 4 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, 66.06 to 53.82. 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 knowledge, where it averages 66.4 against 53.3. The single biggest benchmark swing on the page is HLE, 50.4% to 24%. Qwen3.6-27B does hit back in mathematics, so the answer changes if that is the part of the workload you care about most.
GLM-5 is also the more expensive model on tokens at $1.00 input / $3.20 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen3.6-27B. That is roughly Infinityx on output cost alone. Qwen3.6-27B 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. Qwen3.6-27B gives you the larger context window at 262K, 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 | Qwen3.6-27B |
|---|---|---|---|
| Math | GLM-556.3 | Margin→ 32.9 | Qwen3.6-27B89.2 |
| Knowledge | GLM-566.4 | Margin← 13.1 | Qwen3.6-27B53.3 |
| Coding | GLM-566.3 | Margin→ 11.2 | Qwen3.6-27B77.5 |
| Agentic | GLM-556.2 | Margin→ 3.1 | Qwen3.6-27B59.3 |
| Reasoning | GLM-560.8 | MarginNo overlap | Qwen3.6-27BNot measured |
| Multilingual | GLM-583.1 | MarginNo overlap | Qwen3.6-27BNot measured |
| Multimodal | GLM-5Not measured | MarginNo overlap | Qwen3.6-27B76.7 |
| Inst. Following | GLM-592.6 | MarginNo overlap | Qwen3.6-27BNot measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
HLE
KnowledgeA 50.4%B 24%Winner: GLM-5Δ 26.4HLE: GLM-5 scored 50.4%; Qwen3.6-27B scored 24%. GLM-5 wins this benchmark. - Source ↗
Terminal-Bench 2.0
AgenticA 56.2%B 59.3%Winner: Qwen3.6-27BΔ 3.1Terminal-Bench 2.0: GLM-5 scored 56.2%; Qwen3.6-27B scored 59.3%. Qwen3.6-27B wins this benchmark. - Source ↗
HMMT Feb 2026
MathA 86.4%B 84.3%Winner: GLM-5Δ 2.1HMMT Feb 2026: GLM-5 scored 86.4%; Qwen3.6-27B scored 84.3%. GLM-5 wins this benchmark. - Source ↗
GPQA
KnowledgeA 86%B 87.8%Winner: Qwen3.6-27BΔ 1.8GPQA: GLM-5 scored 86%; Qwen3.6-27B scored 87.8%. Qwen3.6-27B wins this benchmark. - Source ↗
AIME26
MathA 95.8%B 94.1%Winner: GLM-5Δ 1.7AIME26: GLM-5 scored 95.8%; Qwen3.6-27B scored 94.1%. 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 | Qwen3.6-27B | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | GLM-5$1 input / $3.2 output | Qwen3.6-27B$0 input / $0 output | Qwen3.6-27B has the lower combined listed price. |
| Generation speedtokens per second | GLM-574 tok/s | Qwen3.6-27BNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | GLM-51.64 s | Qwen3.6-27BNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | GLM-5200K | Qwen3.6-27B262K | Qwen3.6-27B lists the larger context window. |
Benchmark Deep Dive
AgenticQwen3.6-27B wins18 benchmarks
| Benchmark | GLM-5 | Qwen3.6-27B | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 56.2% | 59.3% | Qwen3.6-27B leads |
| Claw-EvalSource | 57.7% | 72.4% | Qwen3.6-27B leads |
| QwenClawBenchSource | 54.1% | 53.4% | GLM-5 leads |
| 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% | 94.2% | GLM-5 leads |
| CyberGymSource | 43.2% | — | Not comparable |
| APEX-Agents-AASource | 14.5% | — | Not comparable |
| Gert LabsSource | 50.99% | 54.84% | Qwen3.6-27B leads |
| QwenWebBenchSource | — | 1487 | Not comparable |
| AndroidWorldSource | — | 70.3% | Not comparable |
| AA Agentic IndexSource | — | 27.0% | Not comparable |
| GDPval-AASource | — | 32.0% | Not comparable |
| GDPval-AASource | — | 1140 | Not comparable |
CodingQwen3.6-27B wins11 benchmarks
| Benchmark | GLM-5 | Qwen3.6-27B | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 77.8% | 77.2% | GLM-5 leads |
| SWE-bench Verified*Source | 72.8% | — | Not comparable |
| SWE-bench ProSource | 55.1% | 53.5% | GLM-5 leads |
| SWE MultilingualSource | 73.3% | 71.3% | GLM-5 leads |
| SWE-RebenchSource | 62.8% | — | Not comparable |
| React Native EvalsSource | 74.8% | — | Not comparable |
| AA-SciCodeSource | 46.2% | 39.8% | GLM-5 leads |
| Terminal-Bench 2.0Source | — | 59.3% | Not comparable |
| LiveCodeBenchSource | — | 83.9% | Not comparable |
| NL2RepoSource | — | 36.2% | Not comparable |
| AA Coding IndexSource | — | 53.7% | Not comparable |
Reasoning4 benchmarks
KnowledgeGLM-5 wins14 benchmarks
| Benchmark | GLM-5 | Qwen3.6-27B | Result |
|---|---|---|---|
| GPQASource | 86% | 87.8% | Qwen3.6-27B leads |
| GPQA-DSource | 86.0% | — | Not comparable |
| SuperGPQASource | 66.8% | 66% | GLM-5 leads |
| MMLU-ProSource | 85.7% | 86.2% | Qwen3.6-27B leads |
| MMLU-Pro (Arcee)Source | 85.8% | — | Not comparable |
| HLESource | 50.4% | 24% | GLM-5 leads |
| Artificial Analysis Intelligence IndexSource | 39.5% | 37.0% | GLM-5 leads |
| AA-GPQA DiamondSource | 82.0% | 84.2% | Qwen3.6-27B leads |
| AA-HLESource | 27.2% | 21.6% | GLM-5 leads |
| AA-Omniscience IndexSource | 2.0% | -19.8% | GLM-5 leads |
| AA-Omniscience AccuracySource | 26.9% | 19.2% | GLM-5 leads |
| AA-Omniscience Hallucination RateSource | 34.0% | 48.3% | GLM-5 leads |
| MMLU-ReduxSource | — | 93.5% | Not comparable |
| C-EvalSource | — | 91.4% | Not comparable |
MathQwen3.6-27B wins8 benchmarks
| Benchmark | GLM-5 | Qwen3.6-27B | Result |
|---|---|---|---|
| AIME26Source | 95.8% | 94.1% | GLM-5 leads |
| AIME25 (Arcee)Source | 93.3% | — | Not comparable |
| HMMT Feb 2025Source | 97.5% | 93.8% | GLM-5 leads |
| HMMT Nov 2025Source | 96.9% | 90.7% | GLM-5 leads |
| HMMT Feb 2026Source | 86.4% | 84.3% | GLM-5 leads |
| MMAnswerBenchSource | 82.5% | 80.8% | GLM-5 leads |
| FrontierMath v2 (Tiers 1-3)Source | 16.434% | — | Not comparable |
| FrontierMath v2 (Tier 4)Source | 2.100% | — | Not comparable |
Multilingual2 benchmarks
Multimodal17 benchmarks
| Benchmark | GLM-5 | Qwen3.6-27B | Result |
|---|---|---|---|
| Design Arena WebsiteSource | 1278 | — | Not comparable |
| MMMUSource | — | 82.9% | Not comparable |
| MMMU-ProSource | — | 75.8% | Not comparable |
| RealWorldQASource | — | 84.1% | Not comparable |
| DynaMathSource | — | 85.6% | Not comparable |
| MStarSource | — | 81.4% | Not comparable |
| SimpleVQASource | — | 56.1% | Not comparable |
| CharXivSource | — | 78.4% | Not comparable |
| CC-OCRSource | — | 81.2% | Not comparable |
| CountBenchSource | — | 97.8% | Not comparable |
| RefCOCO (avg)Source | — | 92.5% | Not comparable |
| ERQASource | — | 62.5% | Not comparable |
| Video-MME (with subtitle)Source | — | 87.7% | Not comparable |
| VideoMMMUSource | — | 84.4% | Not comparable |
| MLVU (M-Avg)Source | — | 86.6% | Not comparable |
| V*Source | — | 94.7% | Not comparable |
| AA-MMMU-ProSource | — | 74.6% | Not comparable |
Frequently Asked Questions (5)
Which is better, GLM-5 or Qwen3.6-27B?
GLM-5 is ahead on BenchLM's BenchAlign leaderboard, 66.06 to 53.82. The biggest single separator in this matchup is HLE, where the scores are 50.4% and 24%.
Which is better for knowledge tasks, GLM-5 or Qwen3.6-27B?
GLM-5 has the edge for knowledge tasks in this comparison, averaging 66.4 versus 53.3. Inside this category, HLE is the benchmark that creates the most daylight between them.
Which is better for coding, GLM-5 or Qwen3.6-27B?
Qwen3.6-27B has the edge for coding in this comparison, averaging 77.5 versus 66.3. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.
Which is better for math, GLM-5 or Qwen3.6-27B?
Qwen3.6-27B has the edge for math in this comparison, averaging 89.2 versus 56.3. Inside this category, HMMT Nov 2025 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, GLM-5 or Qwen3.6-27B?
Qwen3.6-27B has the edge for agentic tasks in this comparison, averaging 59.3 versus 56.2. Inside this category, Claw-Eval is the benchmark that creates the most daylight between them.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Related Comparisons
Explore More
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.