GLM-5.1 vs Qwen3.6 27B
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Following instructions
Wins 1 of 7 areas
Long documents
GLM-5.1 is the stronger all-rounder.Qwen3.6 27B is cheaper and better at long documents.
Scores updated · 31 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software61GLM-5.16 of 7 tests
- ReasoningHard problems that need careful thinking30GLM-5.13 of 3 tests
- MathCompetition and research-level math31GLM-5.13 of 4 tests
- FactsGetting facts right instead of making them up20GLM-5.12 of 2 tests
- AgentsCarrying out multi-step tasks on its own21GLM-5.12 of 3 tests
- Following instructionsDoing exactly what it is asked10GLM-5.11 of 1 test
- Long documentsFinding answers in very long texts01Qwen3.6 27B1 of 1 test
Long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
Qwen3.6 27B costs 27% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GLM-5.1 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+19.4points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+8.7points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+8.4points ahead
Where Qwen3.6 27B pulls ahead
- Reasons across sets of long documentsAA-LCR+3.6points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+3.1points ahead
- Harvard-MIT high-school math contest problemsHMMT Feb. 2026+1.7points ahead
Every test, side by side
All 31 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGLM-5.1
- Terminal-Bench HardGLM-5.1 by 8.443.234.8+8.4
- SWE-bench ProGLM-5.1 by 4.958.453.5+4.9
- SciCodeGLM-5.1 by 244.842.8+2
- SWE-bench MultilingualGLM-5.1 by 273.371.3+2
- LiveCodeBench v6GLM-5.1 by 1.885.783.9+1.8
- Terminal-Bench 2.1GLM-5.1 by 1.161.860.7+1.1
- SWE-bench VerifiedQwen3.6 27B by 176.277.2+1
AgentsGLM-5.1
- GDPValGLM-5.1 by 6.63124.4+6.6
- τ-Bench V3 · BankingQwen3.6 27B by 3.113.616.7+3.1
- Terminal-Bench 4.0GLM-5.1 by 220+2
ReasoningGLM-5.1
- Humanity's Last ExamGLM-5.1 by 730.123.1+7
- CritPtGLM-5.1 by 3.54.61.1+3.5
- GPQA DiamondGLM-5.1 by 2.686.884.2+2.6
FactsGLM-5.1
- AA-Omniscience · Non-hallucinationGLM-5.1 by 19.470.150.7+19.4
- AA-Omniscience · AccuracyGLM-5.1 by 4.123.719.6+4.1
MathGLM-5.1
- IMOAnswerBenchGLM-5.1 by 383.880.8+3
- FrontierMath Tiers 1-3 (v2)GLM-5.1 by 1.736.835.1+1.7
- HMMT Feb. 2026Qwen3.6 27B by 1.782.684.3+1.7
- AIME 2026GLM-5.1 by 1.295.394.1+1.2
Other results10 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- AA-OmniscienceGLM-5.1 by 20.90.8-20+20.9
- Terminal-Bench 2.0GLM-5.1 by 9.76959.3+9.7
- NL2RepoGLM-5.1 by 6.542.736.2+6.5
- AA Agentic IndexGLM-5.1 by 5.125.220.1+5.1
- AA IntelligenceGLM-5.1 by 4.726.121.4+4.7
- τ²-Bench Telecom (AA run)GLM-5.1 by 3.597.794.2+3.5
- HMMT 2025GLM-5.1 by 3.39490.7+3.3
- HMMT Nov. 2025GLM-5.1 by 3.39490.7+3.3
- Artificial Analysis Coding IndexGLM-5.1 by 2.155.853.7+2.1
- MMLU-Protie8686.2tie
Questions people ask
Which is better, GLM-5.1 or Qwen3.6 27B?
GLM-5.1 wins six of the seven areas where both have results: coding, agents, reasoning, facts, math and following instructions. Qwen3.6 27B wins long documents, and costs 27% less.
Which is better for coding?
GLM-5.1. It wins 6 of the 7 coding tests both models report; Qwen3.6 27B wins 1.
Which is cheaper?
GLM-5.1 costs $1.38 per million input tokens and $4.40 per million output tokens; Qwen3.6 27B costs $0.60 and $3.60. That makes Qwen3.6 27B about 27% cheaper for the same work.
How do you compare the two?
We use the 31 benchmark tests both models have published scores on. The verdict counts the 21 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.