glm-4.5: benchmark scores, price and ranking

glm-4.5 from Z.ai (Zhipu AI),Tsinghua University ranks #351 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.4), based on 4 benchmark results captured up to 2026-10-13.

Organisation: Z.ai (Zhipu AI),Tsinghua University
Type: Open-weight / local model
Open weights: Yes
Released: 2025-07-28 (source: curated)
Cheapest API price: $0.400 in / $1.60 out per million tokens (via deepinfra, observed 2026-10-13T21:47Z)
Aggregate rank: #351 of 743 (score 48.4)

Relative to other models, glm-4.5 is strongest on MMLU-Pro (#15 of 136) and weakest on ALE-Bench (#67 of 79).

Across the 4 benchmarks where at least ten models are scored, glm-4.5 is in the top 10% on 0 and in the top half on 1.

Its cheapest listed input price, $0.400 per million tokens, is lower than that of 53% of the 427 models with a listed price.

glm-4.5 benchmark results

glm-4.5 benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
MMLU-Pro	84.6%	#15 of 136	HuggingFace MMLU-Pro leaderboard	2026-10-13
SWE-bench Verified	54.2%	#40 of 73	SWE-bench Verified leaderboard	2026-10-13
Humanity's Last Exam	8.3%	#15 of 27	Epoch AI Benchmarking Hub	2026-10-13
ALE-Bench	345	#67 of 79	Epoch AI Benchmarking Hub	2026-10-13

About glm-4.5

glm-4.5 is an open-weight model that can run on your own hardware from Z.ai (Zhipu AI),Tsinghua University. It was released on 2025-07-28, according to the developer's announcement. The cheapest API listing we track charges $0.400 per million input tokens and $1.60 per million output tokens, through deepinfra. We hold 4 benchmark results for it across 2 areas, and none is in the top tenth of its board.

Where glm-4.5 stands

On our aggregate score, glm-4.5 is #351 of 743 ranked models, around the middle of the table. Its score is 48.4; the scale is centred on 50, so it sits 1.6 points below the typical model. That beats 53% of the ranked field. Among the 4 ranked Z.ai (Zhipu AI),Tsinghua University models it is #3. The next one up is GLM-4.5-Air at 48.7. The next one down is GLM-4.6 at 45.0.

glm-4.5, benchmark by benchmark

MMLU-Pro puts glm-4.5 at 84.6%, #15 of 136. That beats 89% of the models tested, but 3.5 points short of Darwin-180B-RSI (88.1%). The results either side are DeepSeek R1 (0528) (85.0%) and Step-3.5-Flash (84.4%). It covers 57-subject knowledge + reasoning. The board median is 55.3%, so glm-4.5 is 29.3 points above it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.

glm-4.5 lands below the middle on SWE-bench Verified: 54.2%, #40 of 73, with 53% of the field scoring higher. Claude Opus 4.7 (max) leads at 83.5%; the closest result above is Qwen3-Coder-480B-A35B-Instruct at 55.4%. It covers real GitHub issues resolved. The board median is 56.4%, so glm-4.5 is 2.2 points below it. Source: SWE-bench Verified leaderboard, captured 2026-10-13.

glm-4.5 lands below the middle on Humanity's Last Exam: 8.3%, #15 of 27, with 52% of the field scoring higher. Muse Spark leads at 40.6%; the closest result above is Claude Opus 4 at 10.7%. It covers expert-level questions across domains. The board median is 10.7%, so glm-4.5 is 2.4 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

ALE-Bench is a weak spot. glm-4.5 scores 345, #67 of 79, behind 84% of the models measured. The leader, GPT-6 Astra (max), reaches 2,951. It covers AtCoder heuristic-optimisation contests (performance rating). The board median is 764, so glm-4.5 is 419 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Strengths and weaknesses

Software engineering: uneven, ahead of 30% of the field on average across 2 benchmarks. Its best board here is SWE-bench Verified (#40 of 73) and its weakest is ALE-Bench (#67 of 79). Reasoning, maths and science: solid, ahead of 67% of the field on average across 2 benchmarks. Its best board here is MMLU-Pro (#15 of 136) and its weakest is Humanity's Last Exam (#15 of 27).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so glm-4.5 costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1). Similar scores cost more on mimo-v2.5-pro ($0.435) and Inkling Small (xhigh) ($0.450).

Release history

The Z.ai (Zhipu AI),Tsinghua University model released before it was GLM-4.5-Air on 2025-07-20, which scores 48.7. The next one was GLM-4.6 on 2025-09-30, at 45.0.

Reading these numbers

The 4 scores on this page come from 3 sources: Epoch AI Benchmarking Hub (2), HuggingFace MMLU-Pro leaderboard (1), SWE-bench Verified leaderboard (1). 1 of them come from an official leaderboard and 3 from datasets or third-party runs. With 4 results the aggregate rank can move a lot when one new score arrives.