Loading model evidence.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #3 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `intelligenceIndex`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #58 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `omniscienceAccuracy`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #9 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `omniscienceNonHallucination`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #17 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `ifbench`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #209 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `price1mBlended0To3To1`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #218 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `price1mInputTokens`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #202 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `price1mOutputTokens`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #70 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `medianOutputTokensPerSecond`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #63 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstTokenSeconds`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #120 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `medianTimeToFirstAnswerTokenSeconds`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #65 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `opennessBreakdown.opennessIndex`.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #19 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: overall. Source rank: #25. Votes: 3357. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #21 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: creative_writing. Source rank: #29. Votes: 623. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #15 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: english. Source rank: #17. Votes: 1602. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #19 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: exclude_ties. Source rank: #25. Votes: 2479. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #14 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: hard_prompts. Source rank: #17. Votes: 2148. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #13 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: hard_prompts_english. Source rank: #15. Votes: 1074. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #15 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: instruction_following. Source rank: #19. Votes: 1122. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #15 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: longer_query. Source rank: #19. Votes: 1468. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #28 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: multi_turn. Source rank: #37. Votes: 534. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #13 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: overall. Source rank: #16. Votes: 3357. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #12 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: creative_writing. Source rank: #14. Votes: 623. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #12 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: english. Source rank: #14. Votes: 1602. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #12 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: exclude_ties. Source rank: #15. Votes: 2479. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #15 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: hard_prompts. Source rank: #18. Votes: 2148. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #9 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: hard_prompts_english. Source rank: #11. Votes: 1074. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #15 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: instruction_following. Source rank: #20. Votes: 1122. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #16 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: longer_query. Source rank: #20. Votes: 1468. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #17 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: multi_turn. Source rank: #21. Votes: 534. Organization: zai. License: MIT.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #29 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Category: IF. Tasks scored: 4.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #43 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Category: Language. Tasks scored: 3.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #22 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: paraphrase. Category: IF.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #35 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: simplify. Category: IF.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #40 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: story_generation. Category: IF.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #15 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: summarize. Category: IF.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #40 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: connections. Category: Language.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #27 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: plot_unscrambling. Category: Language.

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #63 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: typos. Category: Language.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #5 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `terminalbenchHard`.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #5 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `scicode`.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #7 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `codingIndex`.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #4 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `agenticIndex`.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #2 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: overall. Source rank: #2. Votes: 1994. Organization: zai. License: MIT.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #2 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: webdev. Source rank: #2. Votes: 1994. Organization: zai. License: MIT.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #5 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: webdev-html. Source rank: #7. Votes: 282. Organization: zai. License: MIT.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #2 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: webdev-react. Source rank: #2. Votes: 1712. Organization: zai. License: MIT.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #6 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: code-migration; provider: Zhipu AI.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #61 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: lcb; provider: Zhipu AI.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #4 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: programbench; provider: Zhipu AI.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #3 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: swebench; provider: Zhipu AI.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #7 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: terminal-bench-2-1; provider: Zhipu AI.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #6 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: vibe-code; provider: Zhipu AI.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #9 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: coding. Source rank: #11. Votes: 913. Organization: zai. License: MIT.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #8 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: coding. Source rank: #12. Votes: 913. Organization: zai. License: MIT.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #1 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Category: Agentic Coding. Tasks scored: 3.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #11 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Category: Coding. Tasks scored: 2.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #4 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: javascript. Category: Agentic Coding.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #4 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: typescript. Category: Agentic Coding.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #3 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: python. Category: Agentic Coding.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #35 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: code_generation. Category: Coding.

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #8 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: code_completion. Category: Coding.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #4 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `hle`.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #9 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `gpqa`.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #6 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `critpt`.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #9 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: proof_bench; provider: Zhipu AI.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #27 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: gpqa; provider: Zhipu AI.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #23 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: mmlu_pro; provider: Zhipu AI.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #13 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Category: Mathematics. Tasks scored: 4.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #30 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Category: Reasoning. Tasks scored: 4.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #30 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: AMPS_Hard. Category: Mathematics.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #17 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: integrals_with_game. Category: Mathematics.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #11 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: math_comp. Category: Mathematics.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #21 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: olympiad. Category: Mathematics.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #33 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: theory_of_mind. Category: Reasoning.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #41 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: zebra_puzzle. Category: Reasoning.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #54 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: spatial. Category: Reasoning.

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #5 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: logic_with_navigation. Category: Reasoning.

Agentic performance on economically valuable work tasks.

Rank #3 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `gdpvalBreakdown.elo`.

Applied legal research tasks.

Rank #4 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_research; provider: Zhipu AI.

Applied professional skills tasks.

Rank #10 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: skillsbench; provider: Zhipu AI.

Weighted model performance across economically relevant Vals tasks.

Rank #5 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: vals_index; provider: Zhipu AI.

Completing legal work with documents, spreadsheets, presentations, and file-system tools.

Rank #3 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Zhipu AI.

Academic legal reasoning tasks.

Rank #23 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: legal_bench; provider: Zhipu AI.

Core financial analyst tasks for agentic models.

Rank #7 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: fabv2; provider: Zhipu AI.

Answer quality on tax questions and responses.

Rank #32 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: tax_eval_v2; provider: Zhipu AI.

Medical billing support and coding tasks.

Rank #26 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: medcode; provider: Zhipu AI.

Administrative documentation support for doctors.

Rank #14 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: medscribe; provider: Zhipu AI.

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #23 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: expert. Source rank: #29. Votes: 340. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #20 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_business_and_management_and_financial_operations. Source rank: #26. Votes: 621. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #21 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_entertainment_and_sports_and_media. Source rank: #29. Votes: 804. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #45 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_legal_and_government. Source rank: #59. Votes: 260. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #4 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_life_and_physical_and_social_science. Source rank: #5. Votes: 555. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #17 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_mathematical. Source rank: #20. Votes: 212. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #9 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_medicine_and_healthcare. Source rank: #11. Votes: 246. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #11 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_software_and_it_services. Source rank: #14. Votes: 1299. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #9 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_writing_and_literature_and_language. Source rank: #12. Votes: 827. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #22 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: expert. Source rank: #28. Votes: 340. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #19 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_business_and_management_and_financial_operations. Source rank: #24. Votes: 621. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #11 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_entertainment_and_sports_and_media. Source rank: #14. Votes: 804. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #35 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_legal_and_government. Source rank: #44. Votes: 260. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #3 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_life_and_physical_and_social_science. Source rank: #4. Votes: 555. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #18 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_mathematical. Source rank: #21. Votes: 212. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #7 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_medicine_and_healthcare. Source rank: #9. Votes: 246. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #7 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_software_and_it_services. Source rank: #10. Votes: 1299. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #7 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: industry_writing_and_literature_and_language. Source rank: #9. Votes: 827. Organization: zai. License: MIT.

Structured data manipulation and table reasoning accuracy.

Rank #24 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Category: Data Analysis. Tasks scored: 3.

Average objective performance across LiveBench's current public category mix.

Rank #7 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Category averages included: 7.

Objective consecutive events score in LiveBench.

Rank #20 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: consecutive_events. Category: Data Analysis.

Objective table join score in LiveBench.

Rank #28 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: tablejoin. Category: Data Analysis.

Objective table reformat score in LiveBench.

Rank #92 · Source label: glm-5.2

Derived from the official LiveBench website leaderboard table. Task: tablereformat. Category: Data Analysis.

It matters when the model must browse, call tools, and recover useful answers from external systems.

Rank #2 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `tau2`.

It checks whether long-context claims survive contact with retrieval, memory, or long-document tasks.

Rank #3 · Source label: GLM-5.2 (max)

Parsed from Artificial Analysis public leaderboard field `lcr`.

It checks whether long-context claims survive contact with retrieval, memory, or long-document tasks.

Rank #13 · Source label: zai/glm-5.2

Parsed from Vals AI BenchmarkView overall scores. Vals slug: corp_fin_v2; provider: Zhipu AI.

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #9 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: russian. Source rank: #11. Votes: 338. Organization: zai. License: MIT.

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #8 · Source label: glm-5.2 (max)

Parsed from Arena leaderboard dataset row `glm-5.2 (max)`. Category: russian. Source rank: #11. Votes: 338. Organization: zai. License: MIT.