Gemini 2.5 Pro vs GLM-4.5
Wins 4 of 6 areas
Coding · Reasoning · Long documents · Following instructions
Wins 1 of 6 areas
Agents
Gemini 2.5 Pro is the stronger all-rounder.GLM-4.5 is cheaper and better at agents.
Scores updated · 24 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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.
- ReasoningHard problems that need careful thinking30Gemini 2.5 Pro3 of 3 tests
- CodingWriting and fixing software21Gemini 2.5 Pro2 of 3 tests
- Long documentsFinding answers in very long texts10Gemini 2.5 Pro1 of 1 test
- Following instructionsDoing exactly what it is asked10Gemini 2.5 Pro1 of 1 test
- AgentsCarrying out multi-step tasks on its own01GLM-4.51 of 2 tests · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
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-4.5 costs 75% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemini 2.5 Pro pulls ahead
- Reasons across sets of long documentsAA-LCR+16.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+14points ahead
- Code for real scientific research problemsSciCode+11.5points ahead
Where GLM-4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+20.8points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+16.5points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+1points ahead
Every test, side by side
All 24 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 2.5 Pro
- SciCodeGemini 2.5 Pro by 11.546.334.8+11.5
- Terminal-Bench HardGemini 2.5 Pro by 4.526.522+4.5
- SWE-bench VerifiedGLM-4.5 by 163.264.2+1
ReasoningGemini 2.5 Pro
- GPQA DiamondGemini 2.5 Pro by 5.483.678.2+5.4
- Humanity's Last ExamGemini 2.5 Pro by 51813+5
- CritPtGemini 2.5 Pro by 2.62.60+2.6
FactsEven
- AA-Omniscience · Non-hallucinationGLM-4.5 by 20.89.129.9+20.8
- AA-Omniscience · AccuracyGemini 2.5 Pro by 143925.1+14
Following instructionsGemini 2.5 Pro
- IFBenchGemini 2.5 Pro by 4.648.744.1+4.6
Other results12 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.
- Artificial Analysis Coding IndexGemini 2.5 Pro by 20.446.726.3+20.4
- AA Agentic IndexGLM-4.5 by 12.73.516.2+12.7
- TAU-bench (retail)GLM-4.5 by 12.76779.7+12.7
- Terminal-BenchGLM-4.5 by 12.225.337.5+12.2
- AA-OmniscienceGemini 2.5 Pro by 11.1-16.4-27.4+11.1
- τ²-Bench Telecom (AA run)Gemini 2.5 Pro by 11.154.143+11.1
- TAU-bench (airline)GLM-4.5 by 10.45060.4+10.4
- AIME 2025GLM-4.5 by 10.38393.3+10.3
- AA IntelligenceGemini 2.5 Pro by 2.21512.8+2.2
- MMLU-ProGemini 2.5 Pro by 1.48684.6+1.4
- LiveCodeBenchGemini 2.5 Pro by 1.374.272.9+1.3
- MATH-500 (EM)tie98.898.2tie
Questions people ask
Which is better, Gemini 2.5 Pro or GLM-4.5?
Gemini 2.5 Pro wins four of the six areas where both have results: coding, reasoning, long documents and following instructions. GLM-4.5 wins agents, and costs 75% less. They are level on facts.
Which is better for coding?
Gemini 2.5 Pro. It wins 2 of the 3 coding tests both models report; GLM-4.5 wins 1.
Which is cheaper?
Gemini 2.5 Pro costs $1.25 per million input tokens and $10.00 per million output tokens; GLM-4.5 costs $0.60 and $2.20. That makes GLM-4.5 about 75% cheaper for the same work.
How do you compare the two?
We use the 24 benchmark tests both models have published scores on. The verdict counts the 12 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 12 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.