Zhipu AIGLM 5.1
GLM 5.1 is a Zhipu AI open-weights model, released 7 Apr 2026. Its best configuration (thinking reasoning effort) ranks #199 of 785 on the BenchLeader Index at 56.6 ±8.0, in the upper half. The ± is the point: 378 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 (74) and lowest in reasoning (48). At $1.99 per million tokens blended it is pricier than most ranked models. Output speed of 63 tokens per second puts it slower than most, with a first answer in 62.3 s. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 7 Apr 2026.stale: no new result in six months
- Blended price
- $1.99/M
- $1.29 in · $4.07 out
- Output speed
- 63 tok/s
- measured by Artificial Analysis
- First answer
- 62 s
- first token 1.63 s
- Context
- 200k
- Full answer
- 70 s
- median, reasoning included
- Cost per run
- $1.02
- one full Intelligence Index run
- Released
- 7 Apr 2026
- Overall index57
- Reasoning48
- Agents & tools52
- Knowledge62
- Instruction following74
- Long context62
- Composite59
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 | 50.7 | #363 | 63 tok/s | 1.72 s | $0.0019 | Agents & tools 65 · Composite 57 · Instruction following 53 · Knowledge 51 · Long context 51 · Maths 39 · Reasoning 41 |
| thinkingbest | 56.6 | #199 | 63 tok/s | 62 s | $0.0017 | Agents & tools 52 · Composite 59 · Instruction following 74 · Knowledge 62 · Long context 62 · Reasoning 48 |
| not stated | 55.9 | #217 | 45 tok/s | 1.63 s | $0.0019 | Agents & tools 49 · Coding 56 · Human preference 65 · Knowledge 54 · Maths 55 · Reasoning 61 |
How its index has moved
8 Sept 2026 to 21 Oct 2026- thinking
- 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 | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| GPQA Diamondeffort not stated | – | – | 89.9%#47 | Epoch AI Benchmarking Hub | |
| SimpleBencheffort not stated | – | – | 55.1%#39 | SimpleBench | |
| Arena Hard Promptseffort not stated | – | – | 1489#50 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 83.9%#152 | 86.8%#106 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 27.9%#169 | 30.1%#150 | – | Artificial Analysis | |
| GPQA Diamond (Vals)not in indexeffort not stated | – | – | 84.5%#50 | Vals AI | |
| CritPt | 0.0%#315 | 4.6%#161 | – | Artificial Analysis | |
| Chess Puzzlesnot in indexeffort not stated | – | – | 20.0%#92 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | thinking | not stated | Source | Trend |
|---|---|---|---|---|
| SWE-bench Verified (Epoch)effort not stated | – | 74.2%#16 | Epoch AI Benchmarking Hub | |
| SciCodeeffort not stated | – | 44.8%#143 | SciCode | |
| WeirdMLeffort not stated | – | 57.1%#55 | WeirdML | |
| Arena Codingeffort not stated | – | 1513#51 | Arena (LMArena) | |
| Arena WebDeveffort not stated | – | 1508#58 | Arena (LMArena) | |
| SciCode (AA)not in index | 44.8%#149 | – | Artificial Analysis | |
| LiveCodeBencheffort not stated | – | 81.4%#60 | Vals AI | |
| SWE-bench (Vals)not in indexeffort not stated | – | 76.4%#35 | Vals AI | |
| Code Migrationnot in indexeffort not stated | – | 25.8%#46 | Vals AI | |
| ProgramBenchnot in indexeffort not stated | – | 0.0%#32 | Vals AI | |
| Vibe Code Bench v1.1not in indexeffort not stated | – | 31.5%#73 | Vals AI | |
| ALE-Benchnot in indexeffort not stated | – | 887.1#67 | Epoch AI Benchmarking Hub | |
| Terminal-Bench 4.0 (AA)not in index | 2.0%#118 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 61.8%#92 | – | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| APEX-Agentseffort not stated | – | – | 40.9%#42 | Mercor | |
| Terminal-Bench Hard | 35.6%#74 | 43.2%#41 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 97.1%#13 | 97.7%#9 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (Vals)effort not stated | – | – | 56.9%#47 | Vals AI | |
| MCP Atlaseffort not stated | – | – | 75.6%#17 | Scale AI SEAL | |
| τ³-Banking (AA) | – | 13.6%#125 | – | Artificial Analysis | |
| ITBench SRE (AA)not in index | – | 40.3%#19 | – | Artificial Analysis | |
| Finance Agent v2not in indexeffort not stated | – | – | 44.8%#56 | Vals AI | |
| Harvey's Legal Agent Benchmark v1not in indexeffort not stated | – | – | 0.0%#61 | Vals AI | |
| Legal Research Benchnot in indexeffort not stated | – | – | 27.9%#54 | Vals AI | |
| Terminal-Bench 2.0 (Vals)not in indexeffort not stated | – | – | 53.9%#16 | Vals AI | |
| ExploitBenchnot in indexeffort not stated | – | – | 18.1%#7 | Epoch AI Benchmarking Hub | |
| GBAEvalnot in indexeffort not stated | – | – | 0.0%#22 | Epoch AI Benchmarking Hub | |
| Vending-Bench 2not in indexeffort not stated | – | – | 5634.4#25 | Andon Labs | |
| AutomationBenchnot in index | – | 20.3%#136 | – | Artificial Analysis | |
| GDP.pdf (AA)not in index | – | 8.4%#142 | – | Artificial Analysis |
Maths
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 24.9%#80 | 36.8%#65 | Epoch AI Benchmarking Hub | |
| OTIS Mock AIMEeffort not stated | – | 93.3%#59 | Epoch AI Benchmarking Hub | |
| ProofBencheffort not stated | – | 22.2%#47 | Vals AI | |
| AIME (Vals)effort not stated | – | 91.9%#21 | Vals AI | |
| AIME 2026effort not statedpossible contamination | – | 95.8%#13 | MathArena | |
| HMMT February 2026effort not statedpossible contamination | – | 89.4%#14 | MathArena | |
| MathArena Apexeffort not statedpossible contamination | – | 11.5%#20 | MathArena | |
| Arena Mathseffort not stated | – | 1476#44 | Arena (LMArena) |
Knowledge
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| SimpleQA Verifiedeffort not stated | – | – | 34.0%#56 | Epoch AI Benchmarking Hub | |
| AA-Omniscience | -22.4#251 | 0.8#128 | – | Artificial Analysis | |
| MMLU-Proeffort not stated | – | – | 86.9%#36 | Vals AI | |
| LegalBencheffort not stated | – | – | 84.4%#33 | Vals AI | |
| CorpFineffort not stated | – | – | 64.5%#37 | Vals AI | |
| TaxEvaleffort not stated | – | – | 71.2%#78 | Vals AI | |
| CaseLaw v2not in indexeffort not stated | – | – | 51.5%#47 | Vals AI | |
| MedCodenot in indexeffort not stated | – | – | 41.6%#51 | Vals AI | |
| MedScribenot in indexeffort not stated | – | – | 72.3%#86 | Vals AI | |
| Public Benefits Bench v1.1not in indexeffort not stated | – | – | 61.8%#28 | Vals AI | |
| AA-Omniscience: accuracynot in index | 25.2%#230 | 23.7%#252 | – | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 36.4%#181 | 70.0%#42 | – | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| IFBench | 52.0%#164 | 76.3%#22 | – | Artificial Analysis | |
| Arena Instruction Followingnot in indexeffort not stated | – | – | 1462#48 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1465#53 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1446#46 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1478#45 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1480#47 | Arena (LMArena) |
Long context
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-LCR | 53.3%#307 | 73.7%#173 | Artificial Analysis |
Composite
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | – | 149.8#61 | Epoch AI Benchmarking Hub | |
| AA Intelligence Index v4.3.2 | 24.2#178 | 26.1#157 | – | Artificial Analysis |
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 |
|---|---|---|---|---|---|---|
| Friendli | 98 tok/s | 0.40 s | $1.40 | $4.40 | 203k | – |
| Venice | 67 tok/s | 1.11 s | $1.40 | $4.40 | 200k | fp8 |
| AtlasCloud | 64 tok/s | 1.26 s | $1.26 | $3.96 | 203k | fp8 |
| GMICloud | 63 tok/s | 1.38 s | $1.40 | $4.40 | 203k | fp8 |
| Baidu Qianfan | 61 tok/s | 0.84 s | $0.965 | $3.03 | 203k | fp8 |
| Alibaba Cloud Int. | 50 tok/s | 1.67 s | $1.33 | $4.18 | 203k | fp8 |
| NovitaAI | 49 tok/s | 2.61 s | $1.38 | $4.40 | 205k | fp8 |
| SiliconFlow | 45 tok/s | 1.99 s | $1.19 | $3.74 | 205k | fp8 |
| StreamLake | 39 tok/s | 2.21 s | $0.966 | $3.04 | 200k | fp8 |
| Chutes | 38 tok/s | 2.34 s | $0.980 | $3.08 | 203k | fp8 |
| Phala | 38 tok/s | 1.66 s | $1.21 | $4.20 | 203k | – |
| Nebius Token Factory | 25 tok/s | 0.49 s | $1.40 | $4.40 | 203k | fp8 |
| Z.ai | 25 tok/s | 8.89 s | $1.40 | $4.40 | 203k | fp8 |
Price history
The price above, per 1M tokens, every day it changed.
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 | $0.0017 | $0.0014 | 1.1 min |
| Summarize a 30-page report | 12,000 / 600 | $0.018 | $0.0087 | 1.2 min |
| Code edit | 6,000 / 1,500 | $0.014 | $0.0092 | 1.4 min |
| Agentic coding session | 60,000 / 4,000 | $0.094 | $0.047 | 2.1 min |
| Structured extraction | 2,000 / 200 | $0.0034 | $0.0018 | 1.1 min |
More from Zhipu AI
See also
Questions people ask
- What are GLM 5.1's benchmark scores?
- As of 24 Oct 2026, GLM 5.1 (thinking effort) scores 56.6 ±8.0 on the BenchLeader Index, strongest in instruction following (74) and weakest in reasoning (48). That rests on 7 independent benchmark results, and every score on this page links to its publisher.
- How does GLM 5.1 rank?
- GLM 5.1 (thinking effort) is #199 of 785 ranked configurations on 24 Oct 2026. 378 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 5.1 cost per 1M tokens?
- As of 24 Oct 2026, GLM 5.1 costs $1.29 per 1M input tokens and $4.07 per 1M output tokens, $1.99 blended (three input tokens per output token). The list price last changed on 6 Oct 2026.
- When was GLM 5.1 released?
- Zhipu AI released GLM 5.1 on 7 Apr 2026. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “GLM 5.1: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/glm-5-1.