GLM-4.7 vs GPT-5.2
Wins 0 of 7 areas
—
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Long documents · Following instructions
GPT-5.2 is the stronger all-rounder.GLM-4.7 is cheaper.
Scores updated · 35 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.
- FactsGetting facts right instead of making them up03GPT-5.23 of 4 tests · 1 tie
- CodingWriting and fixing software13GPT-5.23 of 5 tests · 1 tie
- ReasoningHard problems that need careful thinking13GPT-5.23 of 4 tests
- AgentsCarrying out multi-step tasks on its own12GPT-5.22 of 3 tests
- MathCompetition and research-level math01GPT-5.21 of 1 test
- Long documentsFinding answers in very long texts01GPT-5.21 of 1 test
- Following instructionsDoing exactly what it is asked01GPT-5.21 of 1 test
Math, 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.
GLM-4.7 costs 82% 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-5.2 pulls ahead
- Real work tasks from 44 professionsGDPVal+22.5points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+15.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+15points ahead
Where GLM-4.7 pulls ahead
- Common-sense trick questionsSimpleBench+1.9points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+1.1points ahead
Every test, side by side
All 35 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.2
- Terminal-Bench HardGPT-5.2 by 15.231.847+15.2
- SciCodeGPT-5.2 by 745.152.1+7
- SWE-bench VerifiedGPT-5.2 by 6.273.880+6.2
- LMArena · WebDevGLM-4.7 by 19 rating points14351416+19 rating
- SWE-bench Multilingualtie66.766.7tie
AgentsGPT-5.2
- GDPValGPT-5.2 by 22.525.848.3+22.5
- BrowseCompGPT-5.2 by 13.85265.8+13.8
- τ-Bench V3 · BankingGLM-4.7 by 1.112.211.1+1.1
ReasoningGPT-5.2
- Humanity's Last ExamGPT-5.2 by 10.327.437.7+10.3
- CritPtGPT-5.2 by 9.91.711.6+9.9
- GPQA DiamondGPT-5.2 by 4.485.990.3+4.4
- SimpleBenchGLM-4.7 by 1.947.745.8+1.9
FactsGPT-5.2
- AA-Omniscience · AccuracyGPT-5.2 by 1529.344.3+15
- AA-Omniscience · Non-hallucinationGPT-5.2 by 11.8718.8+11.8
- SimpleQA VerifiedGPT-5.2 by 4.932.237.1+4.9
- Vectara HHEM hallucination ratelower is bettertie11.710.8tie
Other results16 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-OmniscienceGPT-5.2 by 35.5-36.4-0.9+35.5
- AA Agentic IndexGPT-5.2 by 3426.260.2+34
- frontiermath_tier_4_v1GPT-5.2 by 18.8018.8+18.8
- Terminal-Bench 2.0GPT-5.2 by 134154+13
- vectara_avg_summary_lengthGPT-5.2 by 115.7 rating points70.6186.3+115.7 rating
- τ²-Bench Telecom (AA run)GLM-4.7 by 11.195.984.8+11.1
- AA IntelligenceGPT-5.2 by 8.222.230.4+8.2
- AIME 2025GPT-5.2 by 4.395.7100+4.3
- Artificial Analysis Coding IndexGPT-5.2 by 3.445.348.7+3.4
- HMMT Feb. 2025GPT-5.2 by 2.997.1100+2.9
- HLE (with tools)GPT-5.2 by 2.742.845.5+2.7
- MMLU-ProGPT-5.2 by 2.484.386.7+2.4
- HMMT Nov. 2025GPT-5.2 by 2.393.595.8+2.3
- τ²-BenchGLM-4.7 by 1.987.485.5+1.9
- vectara_factual_consistencytie88.389.2tie
- vectara_answer_ratetie99.8100tie
Questions people ask
Which is better, GLM-4.7 or GPT-5.2?
GPT-5.2 wins all seven areas where both have results: coding, agents, reasoning, facts, math, long documents and following instructions. GLM-4.7 wins none, but costs 82% less.
Which is better for coding?
GPT-5.2. It wins 3 of the 5 coding tests both models report; GLM-4.7 wins 1, and 1 is a tie.
Which is cheaper?
GLM-4.7 costs $0.60 per million input tokens and $2.20 per million output tokens; GPT-5.2 costs $1.75 and $14.00. That makes GLM-4.7 about 82% cheaper for the same work.
How do you compare the two?
We use the 35 benchmark tests both models have published scores on. The verdict counts the 19 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 16 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.