z-ai
Z.ai: GLM 5.1
z-ai/glm-5.1
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
z-ai/glm-5.1 is Z.ai's flagship text-in, text-out model aimed at agentic engineering. The vendor claims stronger coding than GLM-5, state-of-the-art SWE-Bench Pro results, and better follow-through on long tasks spanning hundreds of iterations and thousands of tool calls. The OpenRouter catalog lists a 204,800-token context window, optional reasoning, tool calling and structured outputs, and its Design Arena measurements look strongest in data visualization, spreadsheet and slide-generation tasks. Evidence here is limited to the vendor card and one catalog, so treat reported gains as claims pending independent validation.[1][2]
Strengths
- Vendor-positioned for long-horizon agentic coding, with claimed ability to stay productive across hundreds of iterations and thousands of tool calls.[2]
- Catalog data show its strongest OpenRouter Design Arena results in data visualization, spreadsheet and slide categories, plus a large documented context window and optional reasoning.[1]
- Vendor lists MIT-licensed weights and local serving through vLLM, SGLang, xLLM, Transformers and KTransformers.[2]
Limitations
- Only a vendor model card and one marketplace catalog were supplied; benchmark gains are source-reported, so independent validation on your workloads is still needed.[1][2]
- Both sources document text input and output only; support for images, audio or other modalities is unverified.[1][2]
- Reasoning is enabled by default in the catalog but no effort levels are listed, so tunable reasoning behavior is unclear from this evidence.[1]
Where to use it
- Worth piloting for multi-step software engineering such as repository generation or terminal-driven fixes, where the vendor claims its long-horizon agentic behavior helps most.[2]
- Plausible fit for API workflows needing tool calling or structured outputs, per the catalog's supported parameters; confirm behavior in testing.[1]
- Candidate for self-hosted serving via the vendor-documented stacks, subject to your own hardware, cost and latency checks.[2]
Sources
- [1] OpenRouter catalog · Z.ai: GLM 5.1Sources checked: Nov 13, 2026
- [2] Model documentation · Z.ai: GLM 5.1Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks. Unlike previous models built around minute-level interactions, GLM-5.1 can work independently and continuously on...
Specifications
- Model ID
- z-ai/glm-5.1
- Canonical ID
- z-ai/glm-5.1-20260406
- Creator
- z-ai
- Added to OpenRouter
- 07.04.2026
- Context length
- 204,800
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $1.4 / out $4.4
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Coding | BB — next 30% of the cohort | — |
| Agents | BB — next 30% of the cohort | parameters: structured_outputs, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Coding index | 55.8 | BB — next 30% of the cohort | 13.11.2026 |
| Artificial Analysis | Agentic index | 23.9 | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · 3dwin rate 62.6% · rank 13 | 1,336Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · asciiartwin rate 44.5% · rank 48 | 1,142Elo | CC — below the B band | 13.11.2026 |
| Design Arena | models · codecategorieswin rate 53.2% · rank 36 | 1,261Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · datavizwin rate 67% · rank 2 | 1,366Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · gamedevwin rate 54.6% · rank 32 | 1,256Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · svgwin rate 55.1% · rank 25 | 1,209Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · uicomponentwin rate 54.9% · rank 20 | 1,292Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · websitewin rate 55.1% · rank 20 | 1,290Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · agenticgamedevwin rate 49% · rank 19 | 1,111Elo | CC — below the B band | 13.11.2026 |
| Design Arena | agents · agentichtmlslideswin rate 52% · rank 5 | 1,205Elo | —Not rated | 13.11.2026 |
| Design Arena | agents · agenticslideswin rate 54.4% · rank 2 | 1,245Elo | —Not rated | 13.11.2026 |
| Design Arena | agents · agenticslides(html)win rate 51.8% · rank 5 | 1,204Elo | —Not rated | 13.11.2026 |
| Design Arena | agents · agenticslides(python-pptx)win rate 53.3% · rank 3 | 1,240Elo | —Not rated | 13.11.2026 |
| Design Arena | agents · androidnativewin rate 51.9% · rank 22 | 1,173Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · fullstackwin rate 53% · rank 25 | 1,146Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · godotgamedevwin rate 37.8% · rank 34 | 1,040Elo | CC — below the B band | 13.11.2026 |
| Design Arena | agents · htmlslideswin rate 48.9% · rank 16 | 1,066Elo | CC — below the B band | 13.11.2026 |
| Design Arena | agents · mobileappswin rate 49.7% · rank 30 | 1,154Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · pdfwin rate 53.8% · rank 17 | 1,197Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · pptxslideswin rate 53.5% · rank 3 | 1,241Elo | —Not rated | 13.11.2026 |
| Design Arena | agents · python-pptxslideswin rate 54.2% · rank 4 | 1,258Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · spreadsheetwin rate 72.4% · rank 3 | 1,315Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · webappswin rate 50.5% · rank 30 | 1,158Elo | CC — below the B band | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| Chutesfp8 | 202,752 | $0.900 | $2.9 | — | — | 100% |
| StreamLakefp8 | 200,000 | $0.966 | $3.04 | — | — | 100% |
| SiliconFlowfp8 | 204,800 | $1.19 | $3.74 | — | — | 100% |
| AtlasCloudfp8 | 202,752 | $1.26 | $3.96 | — | — | 100% |
| Alibabafp8 | 202,745 | $1.33 | $4.18 | — | — | — |
| Novitafp8 | 204,800 | $1.38 | $4.4 | — | — | — |
| Baidufp8 | 202,752 | $1.4 | $4.4 | — | — | — |
| GMICloudfp8 | 202,752 | $1.4 | $4.4 | — | — | — |
| Friendli | 202,752 | $1.4 | $4.4 | — | — | 100% |
| Z.AIfp8 | 202,752 | $1.4 | $4.4 | — | — | 100% |
| Venicefp8 | 200,000 | $1.4 | $4.4 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.