Zhipu AIGLM 5.3 Flash
Reasoning effort
GLM 5.3 Flash is a Zhipu AI open-weights reasoning model, released 20 Aug 2026. Its best configuration ranks #79 of 775 on the BenchLeader Index at 62.8 ±3.9, in the upper half. The ± is the point: 160 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in maths (70) and lowest in agents & tools (60). At $0.238 per million tokens blended it is cheaper than most ranked models. Output speed of 77 tokens per second puts it slower than most, with a first answer in 28.6 s. It has been measured at 5 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 19 Oct 2026.
- Blended price
- $0.238/M
- $0.150 in · $0.500 out
- Output speed
- 77 tok/s
- measured by Artificial Analysis
- First answer
- 29 s
- first token 0.61 s
- Context
- 1M
- Full answer
- 35 s
- median, reasoning included
- Cost per run
- $0.284
- one full Intelligence Index run
- Released
- 20 Aug 2026
- Overall index63
- Reasoning67
- Coding63
- Agents & tools60
- Maths70
- Knowledge65
- Human preference67
- Multimodal64
- Long context65
- Composite65
Versions
Zhipu AI has shipped 3 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| GLM 5.3 Flashthis page | 20 Aug 2026 | 62.8 | #79 |
| GLM 4.7 Flash | 19 Jan 2026 | 51.4 | #344 |
| GLM-4.5-Flash | 28 Jul 2025 | – | – |
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | – | – | 77 tok/s | 29 s | $0.0002 | Agents & tools 44 |
| low | – | – | 77 tok/s | 29 s | $0.0002 | Reasoning 44 |
| high | – | – | 77 tok/s | 29 s | $0.0002 | Reasoning 55 |
| max | 54.2 | #254 | 77 tok/s | 29 s | $0.0002 | Agents & tools 51 · Coding 52 · Knowledge 59 · Maths 51 · Reasoning 63 |
| not statedbest | 62.8 | #79 | 77 tok/s | 29 s | $0.0002 | Agents & tools 60 · Coding 63 · Composite 65 · Human preference 67 · Knowledge 65 · Long context 65 · Maths 70 · Multimodal 64 · Reasoning 67 |
How its index has moved
8 Sept 2026 to 20 Oct 2026- not stated
- max
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Metadata
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA Openness Indexnot in index | 44.4#27 | Artificial Analysis |
Reasoning
| Benchmark | low | high | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| GPQA Diamond | – | – | 90.2%#44 | – | Epoch AI Benchmarking Hub | |
| Arena Hard Promptseffort not stated | – | – | – | 1500#34 | Arena (LMArena) | |
| LiveBench 2.0 Reasoningnot in indexeffort not stated | – | – | – | 77.6%#42 | LiveBench | |
| LiveBench 1.0 Reasoningnot in indexeffort not stated | – | – | – | 77.6%#57 | LiveBench | |
| GPQA Diamond (AA)not in indexeffort not stated | – | – | – | 91.2%#49 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in indexeffort not stated | – | – | – | 39.9%#85 | Artificial Analysis | |
| GPQA Diamond (Vals)not in index | – | – | 86.4%#41 | – | Vals AI | |
| ARC-AGI-1 | 47.0%#172 | 71.8%#132 | 91.0%#63 | – | ARC Prize | |
| ARC-AGI-2 | 27.9%#127 | 50.1%#103 | 65.8%#68 | – | ARC Prize | |
| CritPteffort not stated | – | – | – | 15.4%#88 | Artificial Analysis | |
| Chess Puzzlesnot in index | – | – | 21.0%#87 | – | Epoch AI Benchmarking Hub | |
| Mystery Game Puzzlesnot in index | – | – | 8.0%#118 | – | Epoch AI Benchmarking Hub | |
| BTF-3not in indexeffort not stated | – | – | – | 14.9%#1 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | low | high | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| SciCodeeffort not stated | – | – | – | 51.6%#93 | SciCode | |
| FrontierCode | – | – | 31.8%#31 | – | Cognition | |
| Arena Codingeffort not stated | – | – | – | 1525#27 | Arena (LMArena) | |
| Arena WebDeveffort not stated | – | – | – | 1607#26 | Arena (LMArena) | |
| LiveBench 2.0 Codingnot in indexeffort not stated | – | – | – | 80.6%#30 | LiveBench | |
| LiveBench 1.0 Codingnot in indexeffort not stated | – | – | – | 79.0%#26 | LiveBench | |
| SciCode (AA)not in indexeffort not stated | – | – | – | 51.6%#92 | Artificial Analysis | |
| LiveCodeBench | – | – | 80.5%#66 | – | Vals AI | |
| IOI | – | – | 52.5%#22 | – | Vals AI | |
| SWE-bench (Vals)not in index | – | – | 92.0%#10 | – | Vals AI | |
| Code Migrationnot in index | – | – | 20.5%#51 | – | Vals AI | |
| ProgramBenchnot in index | – | – | 0.0%#32 | – | Vals AI | |
| Vibe Code Bench 1-100not in index | – | – | 16.0%#17 | – | Vals AI | |
| Vibe Code Bench v1.1not in index | – | – | 30.8%#72 | – | Vals AI | |
| Surface Evolver Benchnot in index | – | – | 52.5%#17 | – | Epoch AI Benchmarking Hub | |
| DeepSWE v1.1not in index | – | – | 63.4%#30 | – | Epoch AI Benchmarking Hub | |
| CursorBenchnot in index | 26.9%#72 | 31.1%#62 | 36.8%#47 | – | Cursor | |
| ALE-Benchnot in index | – | 303.6#117 | – | – | Epoch AI Benchmarking Hub | |
| FrontierSWEnot in index | – | – | 18.1%#17 | – | Epoch AI Benchmarking Hub | |
| Terminal-Bench 4.0 (AA)not in indexeffort not stated | – | – | – | 32.8%#41 | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in indexeffort not stated | – | – | – | 84.3%#30 | Artificial Analysis | |
| FrontierCode 1.1 (AA)not in indexeffort not stated | – | – | – | 41.0%#23 | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | max | not stated | Source | Trend |
|---|---|---|---|---|---|
| Terminal-Bench | 35.8%#52 | – | – | Terminal-Bench | |
| APEX-Agentseffort not stated | – | – | 52.8%#24 | Mercor | |
| Arena Agenteffort not stated | – | – | -1.5#32 | Arena (LMArena) | |
| LiveBench Long Agenticnot in indexeffort not stated | – | – | 28.4%#32 | LiveBench | |
| LiveBench Browsernot in indexeffort not stated | – | – | 84.0%#10 | LiveBench | |
| LiveBench Agentic Codingnot in indexeffort not stated | – | – | 56.8%#25 | LiveBench | |
| Terminal-Bench 2.1 (Vals) | – | 62.9%#38 | – | Vals AI | |
| GDPval-AA v2.1not in indexeffort not stated | – | – | 57.1%#20 | Artificial Analysis | |
| τ³-Banking (AA)effort not stated | – | – | 47.2%#8 | Artificial Analysis | |
| ITBench SRE (AA)not in indexeffort not stated | – | – | 51.2%#4 | Artificial Analysis | |
| Analyst Agent (AA)not in indexeffort not stated | – | – | 25.0%#32 | Artificial Analysis | |
| CyberBenchnot in index | – | 67.9%#27 | – | Vals AI | |
| Finance Agent v2not in index | – | 57.9%#14 | – | Vals AI | |
| Harvey's Legal Agent Benchmarknot in index | – | 10.4%#13 | – | Vals AI | |
| Legal Research Benchnot in index | – | 45.2%#17 | – | Vals AI | |
| SkillsBenchnot in index | – | 40.2%#29 | – | Vals AI | |
| SREBenchnot in index | – | 0.8%#26 | – | Vals AI | |
| Tax Agent Benchnot in index | – | 31.6%#22 | – | Vals AI | |
| Terminal-Bench 4.0 (Vals)not in index | – | 37.4%#14 | – | Vals AI | |
| Terminal-Bench Sciencenot in index | – | 2.9%#32 | – | Vals AI | |
| AutomationBenchnot in indexeffort not stated | – | – | 60.4%#33 | Artificial Analysis | |
| GDP.pdf (AA)not in indexeffort not stated | – | – | 15.4%#95 | Artificial Analysis | |
| Harvey LAB-AA v1.1not in indexeffort not stated | – | – | 0.8%#19 | Harvey | |
| EnterpriseOps-Gymnot in indexeffort not stated | – | – | 33.2%#36 | Artificial Analysis | |
| AA-Briefcase v1.1not in indexeffort not stated | – | – | 1453#37 | Artificial Analysis | |
| Terminal-Bench-Science 0.1 (AA)not in indexeffort not stated | – | – | 4.8%#38 | Artificial Analysis | |
| CWE-Bench (AA)effort not stated | – | – | 55.8%#8 | Artificial Analysis | |
| DeepSec-Bench (AA)effort not stated | – | – | 19.9%#11 | Artificial Analysis | |
| CyberGym end-to-end (AA)effort not stated | – | – | 74.0%#5 | Artificial Analysis |
Maths
| Benchmark | max | not stated | Source | Trend |
|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 55.8%#46 | – | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4 | 17.1%#55 | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 93.9%#56 | – | Epoch AI Benchmarking Hub | |
| ProofBench | 21.0%#49 | – | Vals AI | |
| LiveBench Mathematicsnot in indexeffort not stated | – | 81.2%#59 | LiveBench | |
| Arena Mathseffort not stated | – | 1506#8 | Arena (LMArena) |
Knowledge
| Benchmark | max | not stated | Source | Trend |
|---|---|---|---|---|
| LiveBench 2.0 Data Analysisnot in indexeffort not stated | – | 76.4%#24 | LiveBench | |
| LiveBench Researchnot in indexeffort not stated | – | 70.1%#29 | LiveBench | |
| LiveBench 1.0 Data Analysisnot in indexeffort not stated | – | 76.4%#34 | LiveBench | |
| AA-Omniscienceeffort not stated | – | 7.5#98 | Artificial Analysis | |
| MMLU-Pro | 86.1%#47 | – | Vals AI | |
| LegalBench | 83.9%#45 | – | Vals AI | |
| TaxEval | 75.6%#16 | – | Vals AI | |
| Excel Modeling Benchmarknot in index | 55.9%#47 | – | Vals AI | |
| MedScribenot in index | 88.9%#9 | – | Vals AI | |
| GDP.pdfnot in index | 14.0%#42 | – | Epoch AI Benchmarking Hub | |
| AA-Omniscience: accuracynot in indexeffort not stated | – | 27.5%#200 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in indexeffort not stated | – | 72.4%#34 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LiveBench Languagenot in index | 77.3%#42 | LiveBench | |
| LiveBench Instruction Followingnot in index | 52.8%#67 | LiveBench | |
| Arena Instruction Followingnot in index | 1472#34 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1474#41 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1434#67 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1474#52 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1481#45 | Arena (LMArena) |
Multimodal
| Benchmark | max | not stated | Source | Trend |
|---|---|---|---|---|
| Arena Visioneffort not stated | – | 1294#24 | Arena (LMArena) | |
| LiveBench Designnot in indexeffort not stated | – | 77.2%#22 | LiveBench | |
| MMMU-Pro (Vals)not in index | 86.0%#23 | – | Vals AI | |
| AA-3Dnot in indexeffort not stated | – | 935#16 | Artificial Analysis |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 80.0%#77 | Artificial Analysis | |
| MLCRnot in index | 51.1%#12 | Artificial Analysis |
Composite
| Benchmark | max | not stated | Source | Trend |
|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | 153.1#46 | Epoch AI Benchmarking Hub | |
| LiveBench 2.0effort not stated | – | 70.6%#23 | LiveBench | |
| LiveBench 1.0not in indexeffort not stated | – | 71.6%#53 | LiveBench | |
| AA Intelligence Index v4.3.2effort not stated | – | 41.8#55 | Artificial Analysis | |
| Vals Indexnot in index | 50.9#19 | – | Vals AI | |
| AA Cyber Indexnot in indexeffort not stated | – | 49.9#4 | Artificial Analysis |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Modal | 166 tok/s | 0.47 s | $0.150 | $0.500 | 1.0M | nvfp4 |
| Makora | 127 tok/s | 0.72 s | $0.140 | $0.470 | 1.0M | – |
| Baseten | 118 tok/s | 0.43 s | $0.150 | $0.500 | 1.0M | fp8 |
| Fireworks | 113 tok/s | 0.68 s | $0.225 | $0.750 | 1.0M | – |
| DigitalOcean | 108 tok/s | 0.92 s | $0.131 | $0.438 | 1.0M | – |
| Baseten | 107 tok/s | 0.38 s | $0.150 | $0.500 | 1.0M | fp8 |
| inference.net | 106 tok/s | 0.67 s | $0.120 | $0.388 | 1.0M | fp4 |
| CoreWeave | 105 tok/s | 0.89 s | $0.150 | $0.500 | 1.0M | nvfp4 |
| Friendli | 103 tok/s | 1.61 s | $0.150 | $0.500 | 1.0M | – |
| Prime Intellect | 101 tok/s | 0.96 s | $0.150 | $0.500 | 1.0M | – |
| Fireworks | 95 tok/s | 1.69 s | $0.150 | $0.500 | 1.0M | – |
| Crusoe | 91 tok/s | 0.75 s | $0.150 | $0.500 | 1.0M | fp4 |
| Parasail | 90 tok/s | 0.67 s | $0.188 | $0.625 | 1.0M | fp4 |
| Reka AI | 89 tok/s | 3.10 s | $0.030 | $2.40 | 1.0M | – |
| Together | 85 tok/s | 1.77 s | $0.150 | $0.500 | 1.0M | – |
| Parasail | 75 tok/s | 1.16 s | $0.150 | $0.500 | 1.0M | fp4 |
| inference.net | 72 tok/s | 0.82 s | $0.120 | $0.388 | 1.0M | fp4 |
| Wafer | 66 tok/s | 0.56 s | $0.074 | $0.200 | 1.0M | nvfp4 |
| Venice | 63 tok/s | 8.76 s | $0.150 | $0.500 | 1.0M | – |
| Z.ai | 47 tok/s | 4.40 s | $0.150 | $0.500 | 1.0M | fp8 |
| NovitaAI | 45 tok/s | 3.63 s | $0.084 | $0.280 | 1.0M | fp8 |
| AtlasCloud | 43 tok/s | 4.53 s | $0.150 | $0.500 | 1.0M | fp8 |
| SiliconFlow | 43 tok/s | 8.37 s | $0.150 | $0.500 | 1.0M | fp8 |
| GMICloud | 43 tok/s | 7.93 s | $0.075 | $0.250 | 1.0M | fp8 |
| Inceptron | 41 tok/s | 1.13 s | $0.100 | $0.690 | 1.0M | fp8 |
| StreamLake | 31 tok/s | 3.51 s | $0.074 | $0.246 | 1.0M | fp8 |
| Phala | 27 tok/s | 4.88 s | $0.128 | $0.425 | 1.0M | nvfp4 |
| Relace | 24 tok/s | 1.65 s | $0.075 | $0.500 | 1.0M | – |
| Open Inference | 23 tok/s | 0.99 s | $0.024 | $0.120 | 1.0M | fp4 |
| DeepInfra | 20 tok/s | 1.72 s | $0.150 | $0.500 | 1.0M | fp4 |
| NextBit | 19 tok/s | 2.98 s | $0.120 | $0.400 | 1.0M | fp4 |
| Sail Research | 18 tok/s | 1.57 s | $0.090 | $0.150 | 1.0M | fp4 |
| Sail Research | 17 tok/s | 1.71 s | $0.090 | $0.150 | 1.0M | fp4 |
| NEAR AI | 10 tok/s | 4.29 s | $0.080 | $0.500 | 1.0M | fp8 |
| Morph | 9 tok/s | 5.26 s | $0.164 | $0.650 | 1.0M | fp8 |
Price history
Listed price per 1M tokens over time, as recorded by OpenRouter for the provider with the longest history.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.030 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0002 | $0.0002 | 32.5 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0021 | $0.0010 | 36.4 s |
| Code edit | 6,000 / 1,500 | $0.0017 | $0.0011 | 48.1 s |
| Agentic coding session | 60,000 / 4,000 | $0.011 | $0.0056 | 1.3 min |
| Structured extraction | 2,000 / 200 | $0.0004 | $0.0002 | 31.2 s |
Compare GLM 5.3 Flash with…
- GLM 5.3 Flash vs Claude Opus 5.5
- GLM 5.3 Flash vs Gemini 4 Argon
- GLM 5.3 Flash vs Muse Wonder Full
- GLM 5.3 Flash vs GPT-6 Astra
- GLM 5.3 Flash vs Claude Fable 5.1
- GLM 5.3 Flash vs GPT-6.1 Sol
- GLM 5.3 Flash vs Claude Sonnet 5.5
- GLM 5.3 Flash vs Claude Fable 5
- GLM 5.3 Flash vs Claude Opus 5
- GLM 5.3 Flash vs GPT-5.6 Sol
- GLM 5.3 Flash vs Muse Wonder
- GLM 5.3 Flash vs Muse Spark 1.3
- GLM 5.3 Flash vs GPT-6 Sol
- GLM 5.3 Flash vs GPT-5.5
- GLM 5.3 Flash vs GPT-5.5 Pro
- GLM 5.3 Flash vs Kimi K3
- GLM 5.3 Flash vs GPT-5.4
- GLM 5.3 Flash vs Gemini 3.8 Flash
- GLM 5.3 Flash vs Muse Spark
- GLM 5.3 Flash vs GPT-5.6 Terra
- GLM 5.3 Flash vs MiMo-V2.6-Pro
- GLM 5.3 Flash vs Grok 4.6
- GLM 5.3 Flash vs Grok 4.7
- GLM 5.3 Flash vs GPT-5.3-Codex
- GLM 5.3 Flash vs GLM 5.3
- GLM 5.3 Flash vs Gemini 3.7 Flash
- GLM 5.3 Flash vs GPT-5.4 Pro
- GLM 5.3 Flash vs Claude Opus 4.7
- GLM 5.3 Flash vs Claude Opus 4.8
- GLM 5.3 Flash vs Qwen3.8 Max
- GLM 5.3 Flash vs Gemini 3.5 Flash
- GLM 5.3 Flash vs Muse Spark 1.2
- GLM 5.3 Flash vs Qwen3.8 2.4T A95B
- GLM 5.3 Flash vs DeepSeek V4.1 Flash 1017
- GLM 5.3 Flash vs Muse Spark 1.1
- GLM 5.3 Flash vs Claude Opus 4.6
- GLM 5.3 Flash vs DeepSeek V4 Pro
- GLM 5.3 Flash vs GLM 5.2
- GLM 5.3 Flash vs Qwen3.6 Max
See also
Data as of 21 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “GLM 5.3 Flash: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/glm-5-3-flash, data as of 21 Oct 2026.