Gemini 3 Pro vs GLM-5
Wins 1 of 7 areas
Reasoning
Wins 4 of 7 areas
Coding · Agents · Facts · Following instructions
GLM-5 wins more areas, narrowly.Gemini 3 Pro is better at reasoning.
Scores updated · 40 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 software13GLM-53 of 5 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own13GLM-53 of 4 tests
- FactsGetting facts right instead of making them up12GLM-52 of 3 tests
- Following instructionsDoing exactly what it is asked01GLM-51 of 1 test
- ReasoningHard problems that need careful thinking50Gemini 3 Pro5 of 5 tests
- MathCompetition and research-level math11Even1 each · 1 tie
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
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-5 costs 70% 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 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+56.2points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+16.7points ahead
- Multi-step tasks using many real software toolsMCP Atlas+13.7points ahead
Where Gemini 3 Pro pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+29.5points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+26.2points ahead
- Common-sense trick questionsSimpleBench+23.2points ahead
Every test, side by side
All 40 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGLM-5
- SWE-bench ProGLM-5 by 11.843.355.1+11.8
- SciCodeGemini 3 Pro by 9.956.146.2+9.9
- SWE-bench VerifiedGLM-5 by 1.676.277.8+1.6
- Terminal-Bench HardGLM-5 by 1.541.743.2+1.5
- LMArena · WebDevtie14401434tie
AgentsGLM-5
- BrowseCompGLM-5 by 16.759.275.9+16.7
- MCP AtlasGLM-5 by 13.754.167.8+13.7
- GDPValGLM-5 by 10.434.244.6+10.4
- AA ApexAgentsGemini 3 Pro by 3.918.414.5+3.9
ReasoningGemini 3 Pro
- ARC-AGI-2Gemini 3 Pro by 26.231.14.9+26.2
- SimpleBenchGemini 3 Pro by 23.276.453.2+23.2
- Humanity's Last ExamGemini 3 Pro by 10.439.729.3+10.4
- GPQA DiamondGemini 3 Pro by 8.890.882+8.8
- CritPtGemini 3 Pro by 7.19.12+7.1
FactsGLM-5
- AA-Omniscience · Non-hallucinationGLM-5 by 56.28.564.7+56.2
- AA-Omniscience · AccuracyGemini 3 Pro by 29.555.826.3+29.5
- Vectara HHEM hallucination ratelower is betterGLM-5 by 3.513.610.1+3.5
MathEven
- AIME 2026GLM-5 by 3.791.795.4+3.7
- HMMT Feb. 2026Gemini 3 Pro by 3.686.482.8+3.6
- IMOAnswerBenchtie83.182.5tie
Other results18 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.
- ARC-AGI-1Gemini 3 Pro by 30.37544.7+30.3
- browsecomp_with_context_managerGLM-5 by 16.759.275.9+16.7
- frontiermath_tier_4_v1Gemini 3 Pro by 16.718.82.1+16.7
- AA-OmniscienceGemini 3 Pro by 1515.30.3+15
- AA Agentic IndexGLM-5 by 11.15263.1+11.1
- τ²-Bench Telecom (AA run)GLM-5 by 11.187.198.2+11.1
- τ²-BenchGemini 3 Pro by 8.39889.7+8.3
- HLE (with tools)GLM-5 by 4.645.850.4+4.6
- t2-benchGLM-5 by 4.385.489.7+4.3
- HMMT Nov. 2025GLM-5 by 3.693.396.9+3.6
- vectara_factual_consistencyGLM-5 by 3.586.489.9+3.5
- AIME 2025Gemini 3 Pro by 3.310096.7+3.3
- vectara_avg_summary_lengthGemini 3 Pro by 27.5 rating points101.974.4+27.5 rating
- Artificial Analysis Coding IndexGemini 3 Pro by 2.346.544.2+2.3
- AIME 2026 I (Tools-allowed)GLM-5 by 2.190.692.7+2.1
- Terminal-Bench 2.0GLM-5 by 254.256.2+2
- vectara_answer_ratetie99.499.7tie
- AA Intelligencetie2827.9tie
Questions people ask
Which is better, Gemini 3 Pro or GLM-5?
GLM-5 wins four of the seven areas where both have results: coding, agents, facts and following instructions. Gemini 3 Pro wins reasoning. They are level on math and long documents.
Which is better for coding?
GLM-5. It wins 3 of the 5 coding tests both models report; Gemini 3 Pro wins 1, and 1 is a tie.
Which is cheaper?
Gemini 3 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; GLM-5 costs $1.00 and $3.20. That makes GLM-5 about 70% cheaper for the same work.
How do you compare the two?
We use the 40 benchmark tests both models have published scores on. The verdict counts the 22 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.