z-ai
Z.ai: GLM 4.5V
z-ai/glm-4.5v
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
GLM-4.5V is Z.ai's vision-language model for turning text and image inputs into text. The vendor describes a mixture-of-experts design with 106B total and 12B active parameters, an optional Thinking Mode, and coverage of image, document, chart, GUI, grounding, and video tasks. OpenRouter separately lists only text and image inputs, tool calling, and reasoning-related controls. Vendor materials report leading same-scale benchmark results, but the supplied sources provide no independently measured scores; treat that as an unaudited claim.[1][2]
Strengths
- The vendor describes multimodal reasoning across scene and multi-image analysis, documents and charts, GUI screens, visual grounding, and video; the OpenRouter input list is narrower.[1][2]
- A Thinking Mode switch is described as balancing faster replies against deeper reasoning, while OpenRouter metadata exposes reasoning, tool, and response-format controls.[1][2]
- The Hugging Face repository identifies an MIT license and provides usage paths for Transformers, vLLM, SGLang, and LLaMA-Factory.[2]
Limitations
- OpenRouter lists text output only and text/image inputs only; vendor-claimed video support should be confirmed in the chosen API before use.[1][2]
- The vendor notes remaining issues with frontend HTML formatting and escaping, overthinking or repetition, restated endings, pure-text Q&A, and perception tasks such as counting or identifying people.[2]
- The evidence pack contains catalog data and vendor documentation only, with no independent benchmark or workload measurements; validate accuracy, latency, reliability, and integration behavior on target tasks.[1][2]
Where to use it
- Consider for visual question answering over screenshots, images, charts, or long documents when text-only output is sufficient.[1][2]
- Consider for GUI or desktop-assistant prototypes that read screens, identify icons, or produce bounding boxes; the documentation supplies prompting examples.[2]
- Evaluate for video event recognition or segmentation only after confirming that the serving route accepts video, because the catalog lists no video input.[1][2]
Sources
- [1] OpenRouter catalog · Z.ai: GLM 4.5VSources checked: Nov 13, 2026
- [2] Model documentation · Z.ai: GLM 4.5VSources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
GLM-4.5V is a vision-language foundation model for multimodal agent applications. Built on a Mixture-of-Experts (MoE) architecture with 106B parameters and 12B activated parameters, it achieves state-of-the-art results in video understanding,...
Specifications
- Model ID
- z-ai/glm-4.5v
- Creator
- z-ai
- Added to OpenRouter
- 11.08.2025
- Context length
- 65,536
- Modality
- text+image->text
- Input modalities
- text, image
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.600 / out $1.8
- 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 | input: image |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
No evaluations are supplied for this model.
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) |
|---|---|---|---|---|---|---|
| Novitafp8 | 65,536 | $0.600 | $1.8 | — | — | 100% |
| Z.AIfp8 | 65,536 | $0.600 | $1.8 | — | — | — |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.