ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

GLM-5.2vsStep 3.7 Flash

GLM-5.2 leads 2 of 3 shared benchmarks. GLM-5.2 has the larger context window (1M tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
2 – 1
GLM-5.2 leads
Cheaper per token
list price, input + output
Larger context
GLM-5.2
1M tokens
Provider uptime (30d)
— · —
Z.ai · StepFun
GLM-5.2
Z.ai · released 2026-06-24
text
Context
1M
Max out
131K
Input /1M
$1.4
Output /1M
$4.4
Cached /1M
$0.26
Scores
26 · 9 core
Step 3.7 Flash
StepFun · released 2026-05-29
textvision
Context
256K
Max out
Input /1M
Output /1M
Cached /1M
Scores
19 · 6 core
Quality

Benchmark matrix

BenchmarkGLM-5.2Step 3.7 FlashΔEdge
Reported by both · 3
AIME 202690% matharena95% matharena-5 ptStep 3.7 Flash
HLE (with tools)54.7% 47.2% +7.5 ptGLM-5.2
Terminal-Bench 2.181% 59.6% +21.4 ptGLM-5.2
Only GLM-5.2 reports · 22
GPQA Diamond91.9% epochnot reported
Agents' Last Exam (ALE-CLI)23.8% not reported
AutomationBench v1.0.626.2% not reported
CyberGym77.2% not reported
DeepSWE v1.146.2% not reported
ExploitBench24.4% not reported
ExploitGym 2h29 not reported
ExploitGym 6h39 not reported
FrontierMath Tier 4 v2 (Epoch AI run)29.3% epochnot reported
FrontierMath Tiers 1-3 v2 (Epoch AI run)59.2% epochnot reported
FrontierSWE67.5% not reported
GDPval-AA v21508 not reported
NL2Repo48.9% not reported
OTIS Mock AIME 2024-2025 (Epoch AI run)86.4% epochnot reported
PostTrainBench31.7% not reported
ProgramBench (Almost Solved)9.5% not reported
SimpleQA Verified34.2% epochnot reported
SWE-Bench Pro62.1% not reported
SWE-bench Verified (Epoch AI run)78.7% epochnot reported
SWE-Marathon v1.119.4% not reported
Terminal-Bench 3.04.6% not reported
Toolathlon Verified59.9% not reported
Only Step 3.7 Flash reports · 16
SWE-bench Verifiednot reported76.5%
AA-LCRnot reported63.9%
BrowseCompnot reported75.8%
ClawEval v1.1not reported67.1%
DeepSearchQA Accuracynot reported81.7%
DeepSearchQA F1not reported92.8%
GDPval-Stirrupnot reported1415.8
GDPval-Stirrup Intelligence Indexnot reported45.8
HLE (with tools, text-only)not reported49.7%
ResearchRubricsnot reported71.7%
SimpleVQA (with tools)not reported79.2%
SWE-bench Multilingualnot reported72.4%
SWE-bench Pronot reported56.3%
Toolathlonnot reported49.5%
V* (with Python)not reported95.3%
WorldVQA (with visual search)not reported58.1%
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GLM-5.2 minus Step 3.7 Flash in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecGLM-5.2Step 3.7 FlashEdge
Context window1M tokens256K tokensGLM-5.2
Max output131K tokens
Input price / 1M$1.4
Output price / 1M$4.4
Cached input / 1M$0.26
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$5.8
Modalitiestexttext · visionStep 3.7 Flash
Released2026-06-242026-05-29
Cited benchmark scores2619
Reliability

Provider status

All providers
More matchups

GLM-5.2 vs …

More matchups

Step 3.7 Flash vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run GLM-5.2 and Step 3.7 Flash on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.