z-ai
Z.ai: GLM 4.5
z-ai/glm-4.5
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
z-ai/glm-4.5 is Z.ai's flagship open-weight mixture-of-experts model, described by its developer as purpose-built for agent applications. It has 355 billion total parameters with 32 billion active, offers hybrid thinking and non-thinking modes, supports tool calling, and handles text-only input and output over a context of up to 128k tokens. Weights are released under the MIT license in BF16 and FP8 variants. The vendor reports strong results across its own benchmark suite; independent evidence in this pack is limited to OpenRouter catalog data, so real-world performance should be validated on your workloads.[1][2]
Strengths
- Vendor-reported hybrid reasoning design with separate thinking and non-thinking modes, plus documented tool calling via OpenAI-style tool definitions.[1][2]
- Open weights under the MIT license, with base, hybrid reasoning, and FP8 variants published for commercial use and secondary development.[2]
- Long context window of up to 128k tokens, consistent across the model card and the OpenRouter catalog listing.[1][2]
Limitations
- Self-hosting demands substantial infrastructure: the documentation lists multi-GPU H100 or H200 clusters for full BF16 inference and even larger setups for full context length.[2]
- Performance claims come from the vendor's own evaluations; the only independent data here is OpenRouter catalog and arena listings, so capabilities should be verified against your own tasks.[1][2]
- Text-only input and output, a knowledge cutoff of December 2024, and an OpenRouter-listed expiration date of end of 2026 for this catalog entry.[1]
Where to use it
- Agentic workflows that combine multi-step reasoning with tool and function calling, matching the vendor's stated design goal.[1][2]
- Long-context text tasks such as document analysis or large-codebase work, where the 128k window may help; validate on representative inputs.[1][2]
- Self-hosted deployments where an MIT license, open weights, and FP8 support matter more than minimal hardware footprint.[2]
Sources
- [1] OpenRouter catalog · Z.ai: GLM 4.5Sources checked: Nov 13, 2026
- [2] Model documentation · Z.ai: GLM 4.5Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications. It leverages a Mixture-of-Experts (MoE) architecture and supports a context length of up to 128k tokens. GLM-4.5 delivers significantly...
Specifications
- Model ID
- z-ai/glm-4.5
- Creator
- z-ai
- Added to OpenRouter
- 25.07.2025
- Context length
- 131,072
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.600 / out $2.2
- Knowledge cutoff
- 31.12.2024
- Expiration date
- 31.12.2026
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 | No measurement | — |
| Agents | No measurement | parameters: 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 |
|---|---|---|---|---|
| Design Arena | models · 3dwin rate 59.7% · rank 68 | 1,153Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | models · codecategorieswin rate 54.4% · rank 84 | 1,156Elo | CC — below the B band | 13.11.2026 |
| Design Arena | models · datavizwin rate 53.3% · rank 80 | 1,157Elo | CC — below the B band | 13.11.2026 |
| Design Arena | models · gamedevwin rate 54.4% · rank 81 | 1,135Elo | CC — below the B band | 13.11.2026 |
| Design Arena | models · svgwin rate 50.8% · rank 69 | 1,088Elo | CC — below the B band | 13.11.2026 |
| Design Arena | models · uicomponentwin rate 55% · rank 88 | 1,135Elo | CC — below the B band | 13.11.2026 |
| Design Arena | models · websitewin rate 53.7% · rank 86 | 1,162Elo | 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) |
|---|---|---|---|---|---|---|
| Z.AIfp8 | 131,072 | $0.600 | $2.2 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.