BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases9

GLM-5 vs GPT-5.4

Full profiles: GLM-5 (Zhipu AI) and GPT-5.4 (OpenAI): every benchmark, price and rank.

Updated · Data from 11 sources

As of 29 Oct 2026, GPT-5.4 is ahead of GLM-5 on quality (63.3 vs 60.2 on the BenchLeader Index), but GLM-5 is 3.6× cheaper at $1.55 vs $5.63 per 1M tokens blended. Choose GLM-5 for reasoning (68 vs 66) and fast first answers (46.8 s vs 136.5 s); choose GPT-5.4 for agents & tools (64 vs 58), coding (65 vs 59), and long context (66 vs 63), where it also streams faster.

Key facts (as of 29 Oct 2026)

  • GLM-5 scores 60.2 on the BenchLeader Index and GPT-5.4 scores 63.3 (BenchLeader, 29 Oct 2026).
  • GLM-5 costs $1.55 per 1M tokens blended (list price, 29 Oct 2026).
  • GPT-5.4 costs $5.63 per 1M tokens blended (list price, 29 Oct 2026).
  • GPT-5.4 streams faster than GLM-5 (Artificial Analysis, 7-day median, 29 Oct 2026).
  • Of the 21 benchmarks both models have, GPT-5.4 leads 20 (BenchLeader, 29 Oct 2026).
GLM-5 benchmarks thinking Zhipu AI GPT-5.4 benchmarks xhigh OpenAIOpen in the comparison tool

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingGPT-5.465 vs 59 category score; LiveCodeBench 84.1% vs 81.9%Medium
Agents & tool useGPT-5.464 vs 58 category score; Terminal-Bench Hard 57.6% vs 43.2%Medium
MathsGPT-5.465 vs 63 category score; AIME (Vals) 96.7% vs 91.7%Medium
ReasoningGLM-568 vs 66 category score; GPQA Diamond (AA) 82.0% vs 92.0%Medium
KnowledgeGPT-5.459 vs 57 category score; MMLU-Pro 87.5% vs 86.0%Medium
Long documentsGPT-5.466 vs 63 category score; AA-LCR v1.1 82.0% vs 75.7%Medium
Following instructionsGPT-5.472 vs 71 category score; IFBench 74.0% vs 72.3%Medium
Lowest costGLM-5$1.55 vs $5.63 per 1M blended; agentic coding session $0.073 vs $0.210–
Fastest first answerGLM-546.8 s (GLM-5 (thinking)) vs 136.5 s (GPT-5.4 (xhigh)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • GPT-5.4 (xhigh) leads on quality: 63.3 vs 60.2.
  • GLM-5 (thinking) is stronger in reasoning.
  • GPT-5.4 (xhigh) is stronger in agents & tools, coding, instruction following, knowledge, long context, maths, multimodal.
  • GLM-5 (thinking) is 3.6× cheaper ($1.55 vs $5.63 per 1M blended).
  • GPT-5.4 (xhigh) streams 1.4× faster (94 vs 69 tokens per second).

At a glance

SpecGLM-5GPT-5.4
DeveloperZhipu AIOpenAI
Released11 Feb 20265 Mar 2026
Context window (tokens)204,8001,050,000
Compared atGLM-5 (thinking)GPT-5.4 (xhigh)
Open weightsYesNo
Price in / out / cached input per 1M$1.00 / $3.20 / $0.149$2.50 / $15.00 / $0.250

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryGLM-5 (thinking)GPT-5.4 (xhigh)Leader
Reasoning68.266.1GLM-5
Coding59.164.5GPT-5.4
Agents & tools58.063.6GPT-5.4
Maths63.265.1GPT-5.4
Knowledge57.358.6GPT-5.4
Instruction following70.872.3GPT-5.4
Multimodal–61.1not comparable
Long context62.665.9GPT-5.4

Where sources disagree

Across the 21 benchmarks both have, GLM-5 leads 1, GPT-5.4 leads 20. Every overall index that rates both puts GPT-5.4 ahead.

Overall indexGLM-5 (thinking)GPT-5.4 (xhigh)Leader
BenchLeader Index60.263.3GPT-5.4
Arena Text1457†1475GPT-5.4

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.

MeasureGLM-5 (thinking)GPT-5.4 (xhigh)
Blended price per 1M tokens$1.55$5.63
Input / output per 1M$1.00 / $3.20$2.50 / $15.00
Cached input per 1M$0.149$0.250
Price per task: chat reply$0.0014$0.0055
Price per task: summarize a 30-page report$0.014$0.039
Price per task: code edit$0.011$0.037
Price per task: agentic coding session$0.073$0.210
Price per task: structured extraction$0.0026$0.0080
Output speed69 tok/s94 tok/s
Time to first answer46.8 s136.5 s
Time to a complete answer54.0 s141.9 s

Each model is compared at its best-ranked reasoning-effort setting. GLM-5 (thinking) scores 60.2 and takes 47 s to its first answer token; GLM-5 scores 54.9 in 2 s. GPT-5.4 (xhigh) scores 63.3 and takes 137 s to its first answer token; GPT-5.4 (low) scores 58.9 in 2 s, GPT-5.4 (high) scores 58.8 in 1 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

113 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherGLM-5 (thinking)GPT-5.4 (xhigh)
BenchLeader IndexBenchLeader60.263.3
Agents & tools scoreBenchLeader58.063.6
Coding scoreBenchLeader59.164.5
Instruction following scoreBenchLeader70.872.3
Knowledge scoreBenchLeader57.358.6
Long context scoreBenchLeader62.665.9
Maths scoreBenchLeader63.265.1
Reasoning scoreBenchLeader68.266.1
Multimodal scoreBenchLeader–61.1
Blended price $/MBenchLeader$1.55$5.63
Output speedBenchLeader69 tok/s94 tok/s
Time to first answerBenchLeader46.8 s136.5 s
Context windowBenchLeader204,800 tokens1,050,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub87.8%†93.3%
Humanity's Last ExamScale AI / CAIS–36.2%
SimpleBenchSimpleBench53.2%†–
Arena Hard PromptsArena (LMArena)1478†1487†
LiveBench 1.0 ReasoningLiveBench–88.1%
GPQA Diamond (AA)Artificial Analysis82.0%92.0%
Humanity's Last Exam (AA)Artificial Analysis29.3%43.7%
GPQA Diamond (Vals)Vals AI83.3%91.7%
Kagi LLM BenchmarkKagi LLM Benchmark75.0%63.8%†
ARC-AGI-1ARC Prize44.7%†93.7%
ARC-AGI-2ARC Prize4.9%†74.0%
ARC-AGI-3ARC Prize–0.2%†
CritPtArtificial Analysis2.0%23.4%
Chess PuzzlesEpoch AI Benchmarking Hub–48.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub–37.0%
CL-benchEpoch AI Benchmarking Hub18.7%†27.9%
CL-bench LifeEpoch AI Benchmarking Hub–21.7%
LMCAEpoch AI Benchmarking Hub–52.0%
DTBenchEpoch AI Benchmarking Hub–94.4%
ForecastBenchForecastBench63.7%†63.1%†
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub72.1%†76.9%†
SciCodeSciCode–56.6%
WeirdMLWeirdML48.2%†77.7%
GSO-BenchGSO-Bench–31.4%
Arena CodingArena (LMArena)1498†1511†
Show all 113 benchmarks
MetricPublisherGLM-5 (thinking)GPT-5.4 (xhigh)
Coding
Arena WebDevArena (LMArena)1436†1401†
LiveBench 1.0 CodingLiveBench–77.5%
LiveCodeBenchVals AI81.9%84.1%
SWE-bench (Vals)Vals AI71.4%78.2%
SWE-Bench ProScale AI SEAL–59.1%
SWE-bench Verified (bash only)SWE-bench72.8%†–
SWE-bench Verified (any scaffold)SWE-bench72.8%†–
Code MigrationVals AI–35.0%
ProgramBenchVals AI–0.0%
Vibe Code Bench v1.1Vals AI23.4%67.4%
SWE-Bench Pro (private)Scale AI SEAL–59.1%
SWE Atlas: Codebase QnAScale AI SEAL20.5%†40.8%
SWE Atlas: RefactoringScale AI SEAL24.2%†44.3%
SWE Atlas: Test WritingScale AI SEAL28.7%†44.4%
MirrorCodeEpoch AI Benchmarking Hub–15.6%†
DeepSWE v1.1Epoch AI Benchmarking Hub–51.8%
ALE-BenchEpoch AI Benchmarking Hub765.6†1607†
AlgoTuneAlgoTune–1.9†
Terminal-Bench 2.1 (AA)Artificial Analysis–78.3%
Agents & tools
Terminal-BenchTerminal-Bench52.4%†81.8%†
APEX-AgentsMercor–52.4%†
LiveBench Agentic CodingLiveBench–53.8%
Terminal-Bench HardArtificial Analysis43.2%57.6%
τ²-Bench Telecom (AA)Artificial Analysis98.3%87.1%
MCP AtlasScale AI SEAL–70.6%
HiL-BenchScale AI SEAL–9.7%†
τ²-benchτ²-bench81.0%–
τ³-bench Bankingτ²-bench9.8%39.4%
τ³-Banking (AA)Artificial Analysis–39.6%
APEX-Agents (AA)Artificial Analysis14.4%33.3%
Harvey's Legal Agent Benchmark v1Vals AI–0.0%
SkillsBenchVals AI–51.7%
Terminal-Bench 2.0 (Vals)Vals AI49.4%58.4%†
DeepResearch BenchDeepResearch Bench–35.1%†
PostTrainBenchEpoch AI Benchmarking Hub–19.0%†
METR Time HorizonsMETR–74.3%
GBAEvalEpoch AI Benchmarking Hub–45.1%†
Vending-Bench 2Andon Labs4432.1†6144.2†
DeepSearchQAGoogle–63.7%†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub–78.6%
FrontierMath Tier 4Epoch AI Benchmarking Hub–49.0%
OTIS Mock AIMEEpoch AI Benchmarking Hub80.0%†95.3%
ProofBenchVals AI–56.0%
LiveBench MathematicsLiveBench–94.2%
AIME (Vals)Vals AI91.7%96.7%
AIME 2026MathArena96.7%†99.2%
HMMT February 2026MathArena86.4%†97.7%
USAMO 2026MathArena35.1%†95.2%
MathArena ApexMathArena10.9%†54.2%
Arena MathsArena (LMArena)1444†1460†
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub–45.1%
LiveBench 1.0 Data AnalysisLiveBench–79.3%
MMLU-ProVals AI86.0%87.5%
LegalBenchVals AI84.1%86.0%
CorpFinVals AI62.9%65.3%
TaxEvalVals AI70.0%74.0%
MedQAVals AI94.3%96.1%
PRBench FinanceScale AI SEAL–45.6%†
PRBench LegalScale AI SEAL–44.4%†
MultiNRCScale AI SEAL–58.3%†
CaseLaw v2Vals AI52.5%63.8%
MedCodeVals AI–41.3%
MedScribeVals AI–77.5%
EBR-benchEpoch AI Benchmarking Hub–25.4%
Instruction following
LiveBench LanguageLiveBench–82.6%
LiveBench Instruction FollowingLiveBench–70.2%
IFBenchArtificial Analysis72.3%74.0%
Arena Instruction FollowingArena (LMArena)1447†1461†
Human preference
Arena TextArena (LMArena)1457†1465†
EQ-Bench 4EQ-Bench–1258†
Arena Creative WritingArena (LMArena)1448†1434†
Arena Multi-turnArena (LMArena)1471†1480†
Arena Longer QueriesArena (LMArena)1468†1477†
Multimodal
Arena VisionArena (LMArena)–1292†
MMMU-ProArtificial Analysis–78.4%
VISTAScale AI SEAL–50.9%†
MMMU-Pro (Vals)Vals AI–87.5%
MortgageTaxVals AI–68.3%
VTBScale AI SEAL–29.2%†
Blueprint-Bench 2Epoch AI Benchmarking Hub–27.1%†
Long context
AA-LCR v1.1Artificial Analysis75.7%82.0%
Arena DocumentArena (LMArena)–1470†
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub145.8†157.2†
LiveBench 1.0LiveBench–78.0%
AA Intelligence Index v4.3.2Artificial Analysis27.939.0
AA Progress Index v1Artificial Analysis280.9311.6
Vals RSI IndexVals AI–15.5%
Safety & honesty
MASKScale AI SEAL–89.7%†

Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.

What changed

  • GLM-5 (thinking): cached input price moved from $0.20/M to $0.15/M (25% down).
  • GLM-5 (thinking): BenchLeader Index 60.6 → 60.1.
  • GLM-5 (thinking): 1 new result (AA Progress Index v1); cached input price moved from $0.15/M to $0.20/M (34% up).
  • GPT-5.4 (xhigh): 1 new result (AA Progress Index v1).
  • GLM-5 (thinking): cached input price moved from $0.20/M to $0.15/M (25% down).
  • GLM-5 (thinking): cached input price moved from $0.15/M to $0.20/M (34% up).
  • GLM-5 (thinking): cached input price moved from $0.20/M to $0.15/M (25% down).
  • GLM-5 (thinking): cached input price moved from $0.15/M to $0.20/M (34% up).
  • GPT-5.4 (xhigh): 2 new results (Harvey's Legal Agent Benchmark v1, Vals RSI Index).
  • GLM-5 (thinking): cached input price moved from $0.20/M to $0.15/M (25% down).

Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs

GLM-5 vs GPT-5.4: questions

Is GPT-5.4 better than GLM-5?
GPT-5.4 (xhigh) leads on quality: 63.3 vs 60.2. The BenchLeader Index combines every independent quality benchmark; GPT-5.4 (xhigh) is ahead overall as of 29 Oct 2026, but check the category scores for your use.
GLM-5 vs GPT-5.4: which is cheaper, and will that change?
GLM-5 is cheaper today: $1.55 against $5.63 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is GLM-5 or GPT-5.4 better for coding?
GPT-5.4 scores higher in coding (65 vs 59 on the category index, where 50 is average). On LiveCodeBench, the top shared coding benchmark, it is 81.9% for GLM-5 against 84.1% for GPT-5.4.
Which is faster, GLM-5 or GPT-5.4?
GPT-5.4 streams faster (94 vs 69 output tokens per second, Artificial Analysis 7-day median).
Is GLM-5 or GPT-5.4 better for agents and tool use?
GPT-5.4 scores higher in agentic tasks (64 vs 58 on the category index, where 50 is average). On Terminal-Bench Hard, the top shared agents & tools benchmark, it is 43.2% for GLM-5 against 57.6% for GPT-5.4.
More questions (4)
Which is better for long documents?
GPT-5.4 scores higher on long-context benchmarks (66 vs 63). GPT-5.4 accepts more context: 1,050,000 against 204,800 tokens for GLM-5.
Which has lower latency?
GLM-5 answers first: 46.8 s against 136.5 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on GLM-5 vs GPT-5.4?
Across the 21 benchmarks both have, GLM-5 leads 1, GPT-5.4 leads 20. Every overall index that rates both puts GPT-5.4 ahead.
When was GPT-5.4 released?
GPT-5.4 was released on 5 Mar 2026.