GLM-5: benchmarks, price and rank
GLM-5 from Zhipu is 73rd of 123 rated AI models on Spezia's leaderboard. It costs $1 in, $3.20 out per million tokens.
- 73rd of 123 rated models
- Rating 1,115
- Rated on 8 benchmarks
- Open weights
- Released Feb 11, 2026
- 205,000 tokens of context
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 59th of 69 on ARC-AGI-2
- 72nd of 94 on Epoch Capabilities Index
- 52nd of 68 on SimpleBench
Price: $1 in, $3.20 out per million tokens.
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- BrowseComp75.9%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
27th of 56. The maker's own number, from Zhipu AI, via LLM Stats.
- MCP Atlas67.8%
Multi-step tasks where the model has to pick and combine many real tools, connected through MCP, the standard way agents plug into apps and data.
21st of 33. The maker's own number, from Zhipu AI, via LLM Stats.
- Vending-Bench 2$4,432
The model runs a simulated vending machine business for a year on its own: finding suppliers, ordering stock, setting prices and paying the daily fees.
38th of 60. Independent result from Epoch AI.
- SimpleBench53.2%
Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.
52nd of 68. Independent result from Epoch AI.
- LMArena text1,446
People put a question to two anonymous models and vote for the better answer.
43rd of 107. Independent result from LMArena.
- LMArena WebDev1,435
People ask two anonymous models to build the same web page or app and vote for the better one.
64th of 98. Independent result from LMArena.
- SWE-bench Verified72.1%
Real bug reports and feature requests from popular open-source Python projects. The model has to change the code so the project's hidden tests pass.
Independent result from Epoch AI.
- GPQA Diamond87.8%
Graduate-level questions in biology, physics and chemistry, written by PhDs so that searching the web does not give the answer away.
62nd of 103. Independent result from Epoch AI.
- ARC-AGI-24.9%
Visual puzzles of colored grids, each with a new rule to work out from a few examples.
59th of 69. Independent result from Epoch AI.
- Epoch Capabilities Index146.5
One score built from dozens of benchmarks, set on a common scale so that models tested on different benchmarks can still be compared.
72nd of 94. Independent result from Epoch AI.
Its full profile is in Spezia.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.