Skip to main content

Model comparison

GLM-5.2 vs Soofi S 30B-A3B

Data verified

Head-to-head evidence from 2 shared benchmark results across 1 category. Overall scores shown here use BenchLM's provisional ranking lane.

81/100
Margin
36.0pts
← winning
Soofi Project
45/100
1 category wins0 category wins

Verified leaderboard positions: GLM-5.2 #4; Soofi S 30B-A3B unranked

Evidence parity. GLM-5.2 and Soofi S 30B-A3B share 2 comparable benchmark results. 1 of 8 categories are comparable. 41 results are unique to GLM-5.2; 6 to Soofi S 30B-A3B.

Updated July 15, 2026
Shared results
2
GLM-5.2 only
41
Soofi S 30B-A3B only
6
Comparable categories
1 / 8

Pick GLM-5.2 if you want the stronger benchmark profile. Soofi S 30B-A3B only becomes the better choice if you want the cheaper token bill or you would rather avoid the extra latency and token burn of a reasoning model.

Confidence note. This is a partial-evidence comparison with 2 shared benchmark results across 1 evidence category; 1 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

GLM-5.2 is clearly ahead on the provisional aggregate, 81 to 45. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

GLM-5.2's sharpest advantage is in knowledge, where it averages 59.6 against 49.9. The single biggest benchmark swing on the page is GPQA, 91.2% to 43.4%.

GLM-5.2 is also the more expensive model on tokens at $1.40 input / $4.40 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Soofi S 30B-A3B. That is roughly Infinityx on output cost alone. GLM-5.2 is the reasoning model in the pair, while Soofi S 30B-A3B is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for GLM-5.2 and Soofi S 30B-A3B
CategoryGLM-5.2ΔSoofi S 30B-A3B
KnowledgeGLM-5.259.6Margin 9.7Soofi S 30B-A3B49.9
AgenticGLM-5.281.0MarginNo overlapSoofi S 30B-A3BNot measured
CodingGLM-5.262.1MarginNo overlapSoofi S 30B-A3BNot measured
MathGLM-5.295.9MarginNo overlapSoofi S 30B-A3BNot measured

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · GLM-5.2B · Soofi S 30B-A3B
  1. GPQA

    Knowledge
    Source ↗
    A 91.2%B 43.4%
    Winner: GLM-5.2Δ 47.8
    GPQA: GLM-5.2 scored 91.2%; Soofi S 30B-A3B scored 43.4%. GLM-5.2 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricGLM-5.2Soofi S 30B-A3BComparison
Input / output priceUSD per 1M tokensGLM-5.2$1.4 input / $4.4 outputSoofi S 30B-A3B$0 input / $0 outputSoofi S 30B-A3B has the lower combined listed price.
Generation speedtokens per secondGLM-5.2Not availableSoofi S 30B-A3BNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenGLM-5.2Not availableSoofi S 30B-A3BNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensGLM-5.21MSoofi S 30B-A3B1MListed context windows are equal.

Benchmark Deep Dive

Agentic
BenchmarkGLM-5.2Soofi S 30B-A3BResult
Terminal-Bench 2.0Source 81%Not comparable
MCP AtlasSource 76.8%Not comparable
ToolathlonSource 48.2%Not comparable
AA Agentic IndexSource 43.1%Not comparable
τ²-bench resultsSource 99.1%Not comparable
GDPval-AASource 50.7%Not comparable
GDPval-AASource 1514Not comparable
APEX-Agents-AASource 33.7%Not comparable
ResearchClawBenchSource 20.7%Not comparable
AA BriefcaseSource 1260Not comparable
AA AutomationBenchSource 27.8%Not comparable
AA EnterpriseOps-GymSource 42.7%Not comparable
AA Harvey LABSource 7.5%Not comparable
AA ITBenchSource 42.7%Not comparable
AA Tau3 BankingSource 26.8%Not comparable
Coding
BenchmarkGLM-5.2Soofi S 30B-A3BResult
SWE-bench ProSource 62.1%Not comparable
NL2RepoSource 48.9%Not comparable
Terminal-Bench 2.0Source 81.0%Not comparable
ProgramBenchSource 63.7%Not comparable
cursorBench32Source 55.0%Not comparable
AA Coding IndexSource 68.8%Not comparable
Terminal-Bench HardSource 50.8%Not comparable
AA-SciCodeSource 50.5%Not comparable
AA Terminal-Bench 2.1Source 77.9%Not comparable
HumanEvalSource 73.8%Not comparable
Reasoning
BenchmarkGLM-5.2Soofi S 30B-A3BResult
CritPtSource 20.9%Not comparable
AA-LCRSource 71.3%Not comparable
BBHSource 78.8%Not comparable
DROPSource 66.5%Not comparable
KnowledgeGLM-5.2 wins
BenchmarkGLM-5.2Soofi S 30B-A3BResult
GPQASource 91.2%43.4%GLM-5.2 leads
GPQA-DSource 91.2%43.4%GLM-5.2 leads
HLESource 54.7%Not comparable
HLE w/o toolsSource 40.5%Not comparable
Artificial Analysis Intelligence IndexSource 51.1%Not comparable
AA-GPQA DiamondSource 89.5%Not comparable
AA-HLESource 40.1%Not comparable
AA-Omniscience IndexSource 4.0%Not comparable
AA-Omniscience AccuracySource 25.1%Not comparable
AA-Omniscience Hallucination RateSource 28.1%Not comparable
AA Openness IndexSource 44.4%Not comparable
MMLU-ProSource 51.4%Not comparable
AGIEvalSource 66.9%Not comparable
Math
BenchmarkGLM-5.2Soofi S 30B-A3BResult
AIME26Source 99.2%Not comparable
HMMT Nov 2025Source 94.4%Not comparable
HMMT Feb 2026Source 92.5%Not comparable
MMAnswerBenchSource 91.0%Not comparable
GSM8KSource 86.1%Not comparable
Multimodal
BenchmarkGLM-5.2Soofi S 30B-A3BResult
Design Arena WebsiteSource 1345Not comparable
Inst. Following
BenchmarkGLM-5.2Soofi S 30B-A3BResult
AA-IFBenchSource 73.3%Not comparable
Frequently Asked Questions (2)

Which is better, GLM-5.2 or Soofi S 30B-A3B?

GLM-5.2 is ahead on BenchLM's provisional leaderboard, 81 to 45. The biggest single separator in this matchup is GPQA, where the scores are 91.2% and 43.4%.

Which is better for knowledge tasks, GLM-5.2 or Soofi S 30B-A3B?

GLM-5.2 has the edge for knowledge tasks in this comparison, averaging 59.6 versus 49.9. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 15, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.