Skip to main content

Model comparison

GLM-5 vs Step 3.7 Flash

Data verified

Head-to-head evidence from 19 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

Z.AI
65.98/100
Margin
15.2pts
← winning
50.76/100
1 category wins1 category wins

Verified leaderboard positions: GLM-5 #15; Step 3.7 Flash unranked

BenchAlign evidence: GLM-5 supported; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. GLM-5 and Step 3.7 Flash share 19 comparable benchmark results. 2 of 8 categories are comparable. 31 results are unique to GLM-5; 11 to Step 3.7 Flash.

Updated July 16, 2026
Shared results
19
GLM-5 only
31
Step 3.7 Flash only
11
Comparable categories
2 / 8

Pick GLM-5 if you want the stronger benchmark profile. Step 3.7 Flash only becomes the better choice if agentic is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 19 shared benchmark results across 6 evidence categories; 2 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 is clearly ahead on the provisional aggregate, 63 to 57. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

GLM-5's sharpest advantage is in coding, where it averages 66.3 against 56.3. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 56.2% to 59.5%. Step 3.7 Flash does hit back in agentic, so the answer changes if that is the part of the workload you care about most.

GLM-5 is also the more expensive model on tokens at $1.00 input / $3.20 output per 1M tokens, versus $0.20 input / $1.15 output per 1M tokens for Step 3.7 Flash. That is roughly 2.8x on output cost alone. Step 3.7 Flash is the reasoning model in the pair, while GLM-5 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. Step 3.7 Flash gives you the larger context window at 256K, compared with 200K for GLM-5.

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 and Step 3.7 Flash
CategoryGLM-5ΔStep 3.7 Flash
AgenticGLM-556.2Margin 10.2Step 3.7 Flash66.4
CodingGLM-566.3Margin 10.0Step 3.7 Flash56.3
ReasoningGLM-560.8MarginNo overlapStep 3.7 FlashNot measured
KnowledgeGLM-566.6MarginNo overlapStep 3.7 FlashNot measured
MathGLM-556.3MarginNo overlapStep 3.7 FlashNot measured
MultilingualGLM-583.1MarginNo overlapStep 3.7 FlashNot measured
Inst. FollowingGLM-592.6MarginNo overlapStep 3.7 FlashNot measured

Decisive benchmark drivers

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

More
A · GLM-5B · Step 3.7 Flash
  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 56.2%B 59.5%
    Winner: Step 3.7 FlashΔ 3.3
    Terminal-Bench 2.0: GLM-5 scored 56.2%; Step 3.7 Flash scored 59.5%. Step 3.7 Flash wins this benchmark.
  2. SWE-bench Pro

    Coding
    Source ↗
    A 55.1%B 56.3%
    Winner: Step 3.7 FlashΔ 1.2
    SWE-bench Pro: GLM-5 scored 55.1%; Step 3.7 Flash scored 56.3%. Step 3.7 Flash wins this benchmark.

Operational comparison

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

MetricGLM-5Step 3.7 FlashComparison
Input / output priceUSD per 1M tokensGLM-5$1 input / $3.2 outputStep 3.7 Flash$0.2 input / $1.15 outputStep 3.7 Flash has the lower combined listed price.
Generation speedtokens per secondGLM-574 tok/sStep 3.7 FlashNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenGLM-51.64 sStep 3.7 FlashNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensGLM-5200KStep 3.7 Flash256KStep 3.7 Flash lists the larger context window.

Benchmark Deep Dive

AgenticStep 3.7 Flash wins
BenchmarkGLM-5Step 3.7 FlashResult
Terminal-Bench 2.0Source 56.2%59.5%Step 3.7 Flash leads
Claw-EvalSource 57.7%67.1%Step 3.7 Flash leads
QwenClawBenchSource 54.1%Not comparable
τ³-bench resultsSource 65.6%Not comparable
DeepPlanningSource 14.6%Not comparable
ToolathlonSource 38%49.5%Step 3.7 Flash leads
MCP AtlasSource 31.1%Not comparable
MCP-TasksSource 60.8%Not comparable
WideResearchSource 69.8%Not comparable
τ²-bench resultsSource 98.2%98.5%Step 3.7 Flash leads
CyberGymSource 43.2%Not comparable
APEX-Agents-AASource 14.5%14.8%Step 3.7 Flash leads
Gert LabsSource 50.99%51.57%Step 3.7 Flash leads
BrowseCompSource 75.8%Not comparable
DeepSearchQASource 92.8%Not comparable
GDPval-AASource 25.9%Not comparable
HLE w/ toolsSource 47.2%Not comparable
AA Agentic IndexSource 21.5%Not comparable
GDPval-AASource 1017Not comparable
CodingGLM-5 wins
BenchmarkGLM-5Step 3.7 FlashResult
SWE-bench VerifiedSource 77.8%Not comparable
SWE-bench Verified*Source 72.8%Not comparable
SWE-bench ProSource 55.1%56.3%Step 3.7 Flash leads
SWE MultilingualSource 73.3%Not comparable
SWE-RebenchSource 62.8%Not comparable
React Native EvalsSource 74.8%Not comparable
Terminal-Bench HardSource 43.2%35.6%GLM-5 leads
AA-SciCodeSource 46.2%40.0%GLM-5 leads
Terminal-Bench 2.0Source 59.5%Not comparable
AA Coding IndexSource 39.6%Not comparable
Reasoning
BenchmarkGLM-5Step 3.7 FlashResult
LongBench v2Source 60.8%Not comparable
AI-NeedleSource 63.3%Not comparable
AA-LCRSource 63.3%63.7%Step 3.7 Flash leads
CritPtSource 2.0%2.3%Step 3.7 Flash leads
Knowledge
BenchmarkGLM-5Step 3.7 FlashResult
GPQASource 86%Not comparable
GPQA-DSource 86.0%Not comparable
SuperGPQASource 66.8%Not comparable
MMLU-ProSource 85.7%Not comparable
MMLU-Pro (Arcee)Source 85.8%Not comparable
HLESource 50.4%Not comparable
Artificial Analysis Intelligence IndexSource 39.5%30.3%GLM-5 leads
AA-GPQA DiamondSource 82.0%80.9%GLM-5 leads
AA-HLESource 27.2%19.9%GLM-5 leads
AA-Omniscience IndexSource 2.0%-37.5%GLM-5 leads
AA-Omniscience AccuracySource 26.9%25.4%GLM-5 leads
AA-Omniscience Hallucination RateSource 34.0%84.4%GLM-5 leads
Math
BenchmarkGLM-5Step 3.7 FlashResult
AIME26Source 95.8%Not comparable
AIME25 (Arcee)Source 93.3%Not comparable
HMMT Feb 2025Source 97.5%Not comparable
HMMT Nov 2025Source 96.9%Not comparable
HMMT Feb 2026Source 86.4%Not comparable
MMAnswerBenchSource 82.5%Not comparable
FrontierMath v2 (Tiers 1-3)Source 16.434%Not comparable
FrontierMath v2 (Tier 4)Source 2.100%Not comparable
Multilingual
BenchmarkGLM-5Step 3.7 FlashResult
MMLU-ProXSource 83.1%Not comparable
NOVA-63Source 55.1%Not comparable
Multimodal
BenchmarkGLM-5Step 3.7 FlashResult
Design Arena WebsiteSource 12821218GLM-5 leads
SimpleVQASource 79.2%Not comparable
V*Source 95.3%Not comparable
AA-MMMU-ProSource 75.3%Not comparable
Inst. Following
BenchmarkGLM-5Step 3.7 FlashResult
IFEvalSource 92.6%Not comparable
AA-IFBenchSource 72.3%67.3%GLM-5 leads
Frequently Asked Questions (3)

Which is better, GLM-5 or Step 3.7 Flash?

GLM-5 is ahead on BenchLM's provisional leaderboard, 63 to 57. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 56.2% and 59.5%.

Which is better for coding, GLM-5 or Step 3.7 Flash?

GLM-5 has the edge for coding in this comparison, averaging 66.3 versus 56.3. Inside this category, Terminal-Bench Hard is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, GLM-5 or Step 3.7 Flash?

Step 3.7 Flash has the edge for agentic tasks in this comparison, averaging 66.4 versus 56.2. Inside this category, Toolathlon is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 16, 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.