GLM-4.7 vs MiniMax M2.1
Wins 4 of 7 areas
Coding · Reasoning · Math · Long documents
Wins 2 of 7 areas
Agents · Following instructions
GLM-4.7 wins more areas, narrowly.MiniMax M2.1 is cheaper and better at agents.
Scores updated · 38 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.
- ReasoningHard problems that need careful thinking40GLM-4.74 of 4 tests
- CodingWriting and fixing software41GLM-4.74 of 5 tests
- MathCompetition and research-level math10GLM-4.71 of 1 test
- Long documentsFinding answers in very long texts10GLM-4.71 of 1 test
- AgentsCarrying out multi-step tasks on its own02MiniMax M2.12 of 2 tests
- Following instructionsDoing exactly what it is asked01MiniMax M2.11 of 1 test
- FactsGetting facts right instead of making them up11Even1 each · 1 tie
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.
MiniMax M2.1 costs 46% 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-4.7 pulls ahead
- Olympiad math problems with short answersIMOAnswerBench+21.6points ahead
- Common-sense trick questionsSimpleBench+13points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+8.1points ahead
Where MiniMax M2.1 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+24.5points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+10points ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+5.8points ahead
Every test, side by side
All 38 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGLM-4.7
- SWE-bench VerifiedGLM-4.7 by 6.873.867+6.8
- SWE-bench MultilingualMiniMax M2.1 by 5.866.772.5+5.8
- LMArena · WebDevGLM-4.7 by 51 rating points14351384+51 rating
- SciCodeGLM-4.7 by 4.445.140.7+4.4
- Terminal-Bench HardGLM-4.7 by 331.828.8+3
AgentsMiniMax M2.1
- BrowseCompMiniMax M2.1 by 105262+10
- GDPValMiniMax M2.1 by 3.925.829.7+3.9
ReasoningGLM-4.7
- SimpleBenchGLM-4.7 by 1347.734.7+13
- Humanity's Last ExamGLM-4.7 by 4.227.423.2+4.2
- GPQA DiamondGLM-4.7 by 2.985.983+2.9
- CritPtGLM-4.7 by 1.41.70.3+1.4
FactsEven
- AA-Omniscience · Non-hallucinationMiniMax M2.1 by 24.5731.5+24.5
- AA-Omniscience · AccuracyGLM-4.7 by 8.129.321.2+8.1
- Vectara HHEM hallucination ratelower is bettertie11.711.8tie
Following instructionsMiniMax M2.1
- IFBenchMiniMax M2.1 by 267.969.9+2
Other results21 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.
- HMMT Feb. 2025GLM-4.7 by 26.197.171+26.1
- AA Agentic IndexMiniMax M2.1 by 21.226.247.4+21.2
- HMMT Nov. 2025GLM-4.7 by 19.293.574.3+19.2
- BrowseComp-ZHGLM-4.7 by 18.866.647.8+18.8
- AIME 2025GLM-4.7 by 14.795.781+14.7
- Terminal-BenchMiniMax M2.1 by 14.633.347.9+14.6
- Artificial Analysis Coding IndexGLM-4.7 by 12.545.332.8+12.5
- τ²-Bench Telecom (AA run)GLM-4.7 by 10.595.985.4+10.5
- xbench-DeepSearchGLM-4.7 by 9.352.343+9.3
- xbench-DeepSearch (2025.10)GLM-4.7 by 9.352.343+9.3
- Terminal-Bench 2.0MiniMax M2.1 by 6.94147.9+6.9
- browsecomp_with_context_managerGLM-4.7 by 5.567.562+5.5
- MMLU-ProMiniMax M2.1 by 3.784.388+3.7
- vectara_avg_summary_lengthMiniMax M2.1 by 36.3 rating points70.6106.9+36.3 rating
- AA-OmniscienceMiniMax M2.1 by 3.5-36.4-32.9+3.5
- GAIA (no file)MiniMax M2.1 by 2.461.964.3+2.4
- ResearchRubricsGLM-4.7 by 1.86260.2+1.8
- AA IntelligenceGLM-4.7 by 1.322.220.9+1.3
- vectara_answer_rateGLM-4.7 by 1.399.898.5+1.3
- τ²-Benchtie87.487tie
- vectara_factual_consistencytie88.388.2tie
Questions people ask
Which is better, GLM-4.7 or MiniMax M2.1?
GLM-4.7 wins four of the seven areas where both have results: coding, reasoning, math and long documents. MiniMax M2.1 wins agents and following instructions, and costs 46% less. They are level on facts.
Which is better for coding?
GLM-4.7. It wins 4 of the 5 coding tests both models report; MiniMax M2.1 wins 1.
Which is cheaper?
GLM-4.7 costs $0.60 per million input tokens and $2.20 per million output tokens; MiniMax M2.1 costs $0.30 and $1.20. That makes MiniMax M2.1 about 46% cheaper for the same work.
How do you compare the two?
We use the 38 benchmark tests both models have published scores on. The verdict counts the 17 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 21 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.