Zhipu AIGLM 5.2
GLM 5.2 is a Zhipu AI open-weights reasoning model, released 16 Jun 2026. Its best configuration (max reasoning effort) ranks #86 of 785 on the BenchLeader Index at 62.3 ±3.6, in the upper half. The ± is the point: 167 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 instruction following (72) and lowest in knowledge (54). At $2.15 per million tokens blended it is pricier than most ranked models. Output speed of 91 tokens per second puts it faster than most, with a first answer in 25.2 s. It has been measured at 9 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 22 Oct 2026.
- Blended price
- $2.15/M
- $1.40 in · $4.40 out
- Output speed
- 91 tok/s
- measured by Artificial Analysis
- First answer
- 25 s
- first token 0.46 s
- Context
- 1M
- Full answer
- 31 s
- median, reasoning included
- Cost per run
- $1.47
- one full Intelligence Index run
- Released
- 16 Jun 2026
- Overall index62
- Reasoning70
- Coding62
- Agents & tools63
- Maths57
- Knowledge54
- Instruction following72
- Human preference67
- Long context64
- Composite67
Versions
Zhipu AI has shipped 8 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.
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 | 48.0 | #440 | 175 tok/s | 1.10 s | $0.0019 | Agents & tools 45 · Coding 35 · Composite 55 · Knowledge 59 · Long context 46 · Maths 44 · Reasoning 49 |
| minimal | – | – | 130 tok/s | 0.46 s | $0.0019 | |
| low | – | – | 130 tok/s | 0.46 s | $0.0019 | Maths 56 · Reasoning 63 |
| medium | – | – | 130 tok/s | 0.46 s | $0.0019 | |
| high | – | – | 130 tok/s | 0.46 s | $0.0019 | Coding 62 |
| xhigh | – | – | 130 tok/s | 0.46 s | $0.0019 | |
| maxbest | 62.3 | #86 | 91 tok/s | 25 s | $0.0019 | Agents & tools 63 · Coding 62 · Composite 67 · Human preference 67 · Instruction following 72 · Knowledge 54 · Long context 64 · Maths 57 · Reasoning 70 |
| thinking | – | – | 130 tok/s | 0.46 s | $0.0019 | Reasoning 55 |
| not stated | 52.0 | #329 | 130 tok/s | 0.46 s | $0.0019 | Agents & tools 52 · Coding 49 · Human preference 54 · Knowledge 59 · Maths 47 · Reasoning 53 |
How its index has moved
8 Sept 2026 to 23 Oct 2026- max
- not stated
- no reasoning
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.
Reasoning
| Benchmark | no reasoning | minimal | low | medium | max | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| GPQA Diamond | 71.2%#168 | – | 87.9%#65 | – | 91.9%#29 | – | – | Epoch AI Benchmarking Hub | |
| SimpleBencheffort not stated | – | – | – | – | – | – | 58.8%#32 | SimpleBench | |
| Arena Hard Prompts | – | – | – | – | 1496#40 | – | – | Arena (LMArena) | |
| LiveBench 1.0 Reasoningnot in indexeffort not stated | – | – | – | – | – | – | 78.6%#55 | LiveBench | |
| GPQA Diamond (AA)not in index | 68.6%#311 | – | – | – | 89.5%#74 | – | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 9.8%#329 | – | – | – | 41.1%#80 | – | – | Artificial Analysis | |
| GPQA Diamond (Vals)not in indexeffort not stated | – | – | – | – | – | – | 85.6%#44 | Vals AI | |
| Kagi LLM Benchmark | – | – | – | – | – | 60.0%#50 | 62.6%#44 | Kagi LLM Benchmark | |
| ARC-AGI-1effort not stated | – | – | – | – | – | – | 77.0%#127 | ARC Prize | |
| ARC-AGI-2effort not stated | – | – | – | – | – | – | 22.8%#141 | ARC Prize | |
| CritPt | 3.1%#176 | – | – | – | 20.9%#58 | – | – | Artificial Analysis | |
| Chess Puzzlesnot in index | 19.0%#95 | – | 18.0%#99 | – | 22.0%#84 | – | – | Epoch AI Benchmarking Hub | |
| Mystery Game Puzzlesnot in index | 18.0%#69 | 18.0%#69 | 19.0%#64 | 15.0%#91 | – | – | – | Epoch AI Benchmarking Hub | |
| LMCAnot in index | – | – | – | – | 45.8%#86 | – | – | Epoch AI Benchmarking Hub | |
| DTBenchnot in index | – | – | – | – | 93.6%#63 | – | – | Epoch AI Benchmarking Hub |
Coding
Agents & tools
Maths
| Benchmark | no reasoning | low | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 42.5%#60 | 54.7%#50 | 59.2%#41 | – | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4 | – | – | 29.3%#40 | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 28.9%#209 | 75.6%#126 | 86.4%#93 | – | Epoch AI Benchmarking Hub | |
| ProofBench | – | – | 35.0%#41 | – | Vals AI | |
| LiveBench Mathematicsnot in indexeffort not stated | – | – | – | 89.8%#37 | LiveBench | |
| AIME 2026effort not statedpossible contamination | – | – | – | 90.0%#27 | MathArena | |
| HMMT February 2026effort not statedpossible contamination | – | – | – | 92.4%#13 | MathArena | |
| MathArena Apexeffort not statedpossible contamination | – | – | – | 19.8%#15 | MathArena | |
| Arena Maths | – | – | 1488#33 | – | Arena (LMArena) |
Knowledge
| Benchmark | no reasoning | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| SimpleQA Verified | – | – | 34.2%#55 | – | Epoch AI Benchmarking Hub | |
| LiveBench 1.0 Data Analysisnot in indexeffort not stated | – | – | – | 73.7%#41 | LiveBench | |
| AA-Omniscience | -6.6#169 | – | 4.4#112 | – | Artificial Analysis | |
| MMLU-Proeffort not stated | – | – | – | 86.7%#38 | Vals AI | |
| LegalBencheffort not stated | – | – | – | 84.1%#40 | Vals AI | |
| CorpFineffort not stated | – | – | – | 66.1%#21 | Vals AI | |
| TaxEvaleffort not stated | – | – | – | 73.3%#48 | Vals AI | |
| Excel Modeling Benchmarknot in indexeffort not stated | – | – | – | 61.5%#34 | Vals AI | |
| MedCodenot in indexeffort not stated | – | – | – | 40.8%#58 | Vals AI | |
| MedScribenot in indexeffort not stated | – | – | – | 83.5%#44 | Vals AI | |
| DrugDiscoveryBenchnot in index | – | 36.2%#12 | – | – | Scale AI SEAL | |
| EBR-benchnot in index | – | – | 9.5%#22 | – | Epoch AI Benchmarking Hub | |
| AA-Omniscience: accuracynot in index | 20.3%#304 | – | 24.3%#245 | – | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 66.3%#60 | – | 73.7%#30 | – | Artificial Analysis |
Instruction following
| Benchmark | max | not stated | Source | Trend |
|---|---|---|---|---|
| LiveBench Languagenot in indexeffort not stated | – | 76.2%#45 | LiveBench | |
| LiveBench Instruction Followingnot in indexeffort not stated | – | 62.3%#54 | LiveBench | |
| IFBench | 73.3%#41 | – | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1469#38 | – | Arena (LMArena) |
Human preference
| Benchmark | max | not stated | Source | Trend |
|---|---|---|---|---|
| Arena Text | 1475#37 | – | Arena (LMArena) | |
| EQ-Bench 4effort not stated | – | 1222#13 | EQ-Bench | |
| Arena Creative Writingnot in index | 1453#35 | – | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1475#49 | – | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1486#33 | – | Arena (LMArena) |
Long context
| Benchmark | no reasoning | max | Source | Trend |
|---|---|---|---|---|
| AA-LCR | 42.3%#350 | 78.3%#112 | Artificial Analysis | |
| MLCRnot in index | – | 12.8%#93 | Artificial Analysis |
Composite
| Benchmark | no reasoning | max | not stated | Source | Trend |
|---|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | – | 151.9#51 | Epoch AI Benchmarking Hub | |
| LiveBench 1.0not in indexeffort not stated | – | – | 73.2%#46 | LiveBench | |
| AA Intelligence Index v4.3.2 | 22.4#201 | 33.7#103 | – | 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 |
|---|---|---|---|---|---|---|
| Decart | 417 tok/s | 0.47 s | $1.20 | $8.00 | 1.0M | fp4 |
| Nebius Token Factory | 212 tok/s | 0.61 s | $1.40 | $4.40 | 1.0M | fp4 |
| CoreWeave | 188 tok/s | 0.74 s | $0.620 | $2.42 | 1.0M | fp4 |
| Wafer | 155 tok/s | 0.38 s | $0.034 | $7.00 | 1.0M | fp4 |
| Wafer | 138 tok/s | 0.35 s | $0.190 | $8.00 | 1.0M | fp4 |
| Venice | 136 tok/s | 1.48 s | $1.40 | $4.40 | 1M | fp8 |
| Together | 132 tok/s | 0.31 s | $1.40 | $4.40 | 1.0M | – |
| Baseten | 130 tok/s | 0.36 s | $2.10 | $6.60 | 1.0M | fp8 |
| Inceptron | 122 tok/s | 0.43 s | $0.990 | $1.19 | 1.0M | fp4 |
| Baseten | 121 tok/s | 0.35 s | $2.10 | $6.60 | 1.0M | fp8 |
| Ambient | 107 tok/s | 0.72 s | $0.600 | $2.00 | 203k | fp8 |
| Friendli | 101 tok/s | 0.52 s | $1.40 | $4.40 | 1.0M | – |
| Baseten | 100 tok/s | 1.42 s | $1.40 | $4.40 | 1.0M | fp8 |
| SiliconFlow | 100 tok/s | 1.05 s | $0.700 | $2.20 | 1.0M | fp8 |
| Mistral (nvfp4) | 95 tok/s | 1.15 s | $1.40 | $4.40 | 1.0M | nvfp4 |
| DigitalOcean | 90 tok/s | 1.21 s | $0.700 | $2.20 | 1.0M | – |
| Parasail | 89 tok/s | 0.55 s | $1.40 | $4.40 | 262k | fp4 |
| Baidu Qianfan (fp4) | 80 tok/s | 0.85 s | $0.619 | $2.17 | 1.0M | fp4 |
| Alibaba Cloud Int. | 79 tok/s | 0.77 s | $2.31 | $7.26 | 1M | fp8 |
| Cloudflare | 72 tok/s | 0.73 s | $0.190 | $8.00 | 262k | – |
| Alibaba Cloud Int. | 66 tok/s | 0.99 s | $0.966 | $3.04 | 1M | fp8 |
| Baidu Qianfan (fp8) | 65 tok/s | 1.00 s | $0.189 | $0.594 | 1.0M | fp8 |
| Baseten | 60 tok/s | 0.94 s | $1.40 | $4.40 | 1.0M | fp8 |
| Z.ai | 59 tok/s | 2.88 s | $1.40 | $4.40 | 1.0M | fp8 |
| StreamLake | 52 tok/s | 1.09 s | $0.280 | $0.880 | 1.0M | fp8 |
| DeepInfra | 45 tok/s | 2.83 s | $0.450 | $1.44 | 1.0M | fp4 |
| DeepInfra | 44 tok/s | 1.14 s | $0.563 | $1.80 | 1.0M | fp4 |
| Morph | 42 tok/s | 0.78 s | $0.439 | $4.40 | 1.0M | fp8 |
| NovitaAI | 39 tok/s | 1.85 s | $0.650 | $2.04 | 1.0M | fp8 |
| AtlasCloud | 34 tok/s | 3.58 s | $0.938 | $2.95 | 1.0M | fp8 |
| Phala | 31 tok/s | 2.44 s | $1.26 | $3.00 | 1.0M | fp8 |
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 $0.260 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0019 | $0.0015 | 28.5 s |
| Summarize a 30-page report | 12,000 / 600 | $0.019 | $0.0092 | 31.8 s |
| Code edit | 6,000 / 1,500 | $0.015 | $0.0099 | 41.7 s |
| Agentic coding session | 60,000 / 4,000 | $0.102 | $0.050 | 1.2 min |
| Structured extraction | 2,000 / 200 | $0.0037 | $0.0020 | 27.4 s |
More from Zhipu AI
Compare GLM 5.2 with…
- GLM 5.2 vs Claude Opus 5.5
- GLM 5.2 vs Gemini 4 Argon
- GLM 5.2 vs Muse Wonder Full
- GLM 5.2 vs Claude Fable 5.1
- GLM 5.2 vs GPT-6 Astra
- GLM 5.2 vs GPT-6.1 Sol
- GLM 5.2 vs Claude Fable 5
- GLM 5.2 vs Claude Sonnet 5.5
- GLM 5.2 vs Claude Opus 5
- GLM 5.2 vs GPT-5.6 Sol
- GLM 5.2 vs Muse Wonder
- GLM 5.2 vs Muse Spark 1.3
- GLM 5.2 vs Kimi K3.1 Pro
- GLM 5.2 vs GPT-6 Sol
- GLM 5.2 vs GPT-5.5
- GLM 5.2 vs GPT-5.5 Pro
- GLM 5.2 vs GPT-5.4
- GLM 5.2 vs Kimi K3
- GLM 5.2 vs Gemini 3.8 Flash
- GLM 5.2 vs GPT-5.6 Terra
- GLM 5.2 vs Muse Spark
- GLM 5.2 vs Grok 4.6
- GLM 5.2 vs MiMo-V2.6-Pro
- GLM 5.2 vs DeepSeek V4.1 Flash 1017
- GLM 5.2 vs Grok 4.7
- GLM 5.2 vs GPT-5.3-Codex
- GLM 5.2 vs GLM 5.3
- GLM 5.2 vs Gemini 3.7 Flash
- GLM 5.2 vs Qwen3.8 Max
- GLM 5.2 vs Claude Opus 4.7
- GLM 5.2 vs Claude Opus 4.8
- GLM 5.2 vs GPT-5.4 Pro
- GLM 5.2 vs Muse Spark 1.2
- GLM 5.2 vs Gemini 3.5 Flash
- GLM 5.2 vs Qwen3.8 2.4T A95B
- GLM 5.2 vs Claude Opus 4.6
- GLM 5.2 vs GLM 5.3 Flash
- GLM 5.2 vs Muse Spark 1.1
- GLM 5.2 vs Qwen3.6 Max
See also
Questions people ask
- What are GLM 5.2's benchmark scores?
- As of 24 Oct 2026, GLM 5.2 (max effort) scores 62.3 ±3.6 on the BenchLeader Index, top 100, strongest in instruction following (72) and weakest in knowledge (54). That rests on 23 independent benchmark results, and every score on this page links to its publisher.
- How does GLM 5.2 rank?
- GLM 5.2 (max effort) is #86 of 785 ranked configurations on 24 Oct 2026 (top 100). 167 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does GLM 5.2 cost per 1M tokens?
- As of 24 Oct 2026, GLM 5.2 costs $1.40 per 1M input tokens and $4.40 per 1M output tokens, $2.15 blended (three input tokens per output token).
- Is GLM 5.2 better than Qwen3.6 Max?
- As of 24 Oct 2026, GLM 5.2 (max) and Qwen3.6 Max (max) are neck and neck on quality (62.3 vs 62.3 on the BenchLeader Index, inside the ±4.7 margin of error), and GLM 5.2 (max) costs 1.4× less, which makes it the better buy for most workloads. The head-to-head page has the benchmark-by-benchmark detail.
- When was GLM 5.2 released?
- Zhipu AI released GLM 5.2 on 16 Jun 2026. BenchLeader has results for it at 9 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “GLM 5.2: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/glm-5-2.