GLM-4.6 vs GPT-5
Wins 0 of 7 areas
—
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Long documents · Following instructions
GPT-5 is the stronger all-rounder.GLM-4.6 is cheaper.
Scores updated · 45 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 software17GPT-57 of 8 tests
- AgentsCarrying out multi-step tasks on its own03GPT-53 of 3 tests
- ReasoningHard problems that need careful thinking03GPT-53 of 3 tests
- Following instructionsDoing exactly what it is asked02GPT-52 of 2 tests
- FactsGetting facts right instead of making them up12GPT-52 of 3 tests
- MathCompetition and research-level math01GPT-51 of 1 test
- Long documentsFinding answers in very long texts01GPT-51 of 1 test
Math and long documents 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.
GLM-4.6 costs 75% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where GPT-5 pulls ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+32.1points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+29.7points ahead
- Reasons across sets of long documentsAA-LCR+24.2points ahead
Where GLM-4.6 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+14.2points ahead
Every test, side by side
All 45 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5
- SWE-bench ProGPT-5 by 32.19.741.8+32.1
- Terminal-Bench 2.1GLM-4.6 by 14.249.435.2+14.2
- LMArena · WebDevGPT-5 by 77 rating points13411418+77 rating
- Terminal-Bench HardGPT-5 by 7.62532.6+7.6
- SWE-bench VerifiedGPT-5 by 6.96874.9+6.9
- SciCodeGPT-5 by 4.538.442.9+4.5
- LiveCodeBench v6GPT-5 by 4.282.887+4.2
- SWE-bench MultilingualGPT-5 by 1.553.855.3+1.5
AgentsGPT-5
- BrowseCompGPT-5 by 9.845.154.9+9.8
- τ-Bench V3 · BankingGPT-5 by 8.713.422.1+8.7
- GDPValGPT-5 by 81321+8
ReasoningGPT-5
- Humanity's Last ExamGPT-5 by 1414.528.5+14
- GPQA DiamondGPT-5 by 7.47885.4+7.4
- CritPtGPT-5 by 4.61.15.7+4.6
FactsGPT-5
- AA-Omniscience · AccuracyGPT-5 by 13.426.940.3+13.4
- AA-Omniscience · Non-hallucinationGPT-5 by 11.95.917.8+11.9
- Vectara HHEM hallucination ratelower is betterGLM-4.6 by 5.69.515.1+5.6
Following instructionsGPT-5
- IFBenchGPT-5 by 29.743.473.1+29.7
- Multi-ChallengeGPT-5 by 14.754.969.6+14.7
Other results24 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.
- FinSearchComp-globalGPT-5 by 34.729.263.9+34.7
- AA-OmniscienceGPT-5 by 33.2-41.9-8.7+33.2
- τ²-Bench Telecom (AA run)GPT-5 by 14.370.584.8+14.3
- BrowseComp-ZHGPT-5 by 13.549.563+13.5
- ArtifactsBenchGPT-5 by 13.259.873+13.2
- Terminal-Bench 2.0GPT-5 by 10.724.535.2+10.7
- frontiermath_tier_4_v1GPT-5 by 10.42.112.5+10.4
- vectara_avg_summary_lengthGPT-5 by 85.5 rating points77.2162.7+85.5 rating
- Artificial Analysis Coding IndexGLM-4.6 by 845.837.8+8
- AA Agentic IndexGPT-5 by 7.918.626.5+7.9
- xbench-DeepSearchGPT-5 by 7.87077.8+7.8
- τ²-BenchGPT-5 by 7.275.282.4+7.2
- vectara_factual_consistencyGLM-4.6 by 5.690.584.9+5.6
- vectara_answer_rateGPT-5 by 5.494.599.9+5.4
- HLE (with tools)GPT-5 by 4.830.435.2+4.8
- HMMT 2025GPT-5 by 4.688.793.3+4.6
- AA IntelligenceGPT-5 by 4.518.523+4.5
- GAIA (text only)GPT-5 by 4.571.976.4+4.5
- LiveCodeBenchGPT-5 by 479.583.5+4
- MMLU-ProGPT-5 by 3.983.287.1+3.9
- Terminal-BenchGPT-5 by 3.340.543.8+3.3
- HMMT Nov. 2025GLM-4.6 by 2.591.789.2+2.5
- HMMT Feb. 2025tie89.288.3tie
- AIME 2025tie93.994.6tie
Questions people ask
Which is better, GLM-4.6 or GPT-5?
GPT-5 wins all seven areas where both have results: coding, agents, reasoning, facts, math, long documents and following instructions. GLM-4.6 wins none, but costs 75% less.
Which is better for coding?
GPT-5. It wins 7 of the 8 coding tests both models report; GLM-4.6 wins 1.
Which is cheaper?
GLM-4.6 costs $0.57 per million input tokens and $2.20 per million output tokens; GPT-5 costs $1.25 and $10.00. That makes GLM-4.6 about 75% cheaper for the same work.
How do you compare the two?
We use the 45 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 24 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.