GLM 4.5V
z-ai/glm-4.5v- Z.ai
- Family · GLM 4.5V
- Released 11 Aug 2025
- Open weights
- Text · Image → Text
- Tools · Vision
Not on OSMEXO yet
Spec & list price
Context64K
Max output16K
Input / 1M$0.60
Output / 1M$1.803× input
Scores · standing
Intelligence7.6#324/490
Usage
Tokens / day18.1M▲ 18% 7d
Usage rank#377
Requests / day7.4K
Tool calls / req0.09
Per request · 30 days
Avg input3.8K0.3× median
Avg output4710.7× median
Cache hit26%0.4× median
Reasoning share65%1.2× median
Tool-call failures3.2%1.7× median
Consumption
Peak 52.2M 26 SepLow 5.2M 31 OctRank today #377
News & trends
News · latest from Z.ai
- Hermes and REA and GLM 5.3 = Reverse engineering TikTok app for less than $20scraping.club · 29 Oct
- GLM-5.3 Sometimes Considers How Train-Time Actions Shape Future Behaviorlesswrong.com · 27 Oct
- Cross-request KV cache exposure in the GLM-5.3 Workers AI model (155 responses)blog.cloudflare.com · 26 Oct
- ReplaySSM: Rethinking State Storage for Faster Hybrid-Model DecodingNVIDIA · 12 Oct
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across 2 providers, latest day
Output speed#213/33450 tok/s1.3× slower than median
Fastest provider54 tok/sbest of 2 providers
First token#2281.68 s1.6× longer than median
#29 cheapest of 40 with a similar score
Value
Intelligence Index
Blended price per 1M tokens (log)
Technical details
- Tokenizer
- Other
- Knowledge cutoff
- 2024-12-31
- License
- MIT
- Parameters
- 108B
- Moderation
- none
- Downloads · likes
- 19.3K · 726
Request parameters 14
- tools
- tool_choice
- response_format
- reasoning
- include_reasoning
- temperature
- top_p
- top_k
- max_tokens
- stop
+4 more
- seed
- frequency_penalty
- presence_penalty
- repetition_penalty
When to use
Best for
- Image understanding in chat apps
- Reasoning over charts and diagrams
- Tool-calling multimodal agents
- Self-hosted vision assistants
Think twice for
- Long document analysis
- Video understanding
- New projects (newer vision models exist)
- Strict structured-output pipelines
Research papers
- Benchmarking color judgment in multimodal LLMs with CIE-computable ground truth29 Oct
- True (VIS) Lies: Analyzing How Generative AI Recognizes Intentionality, Rhetoric, and Misleadingness in Visualization Lies1 Apr
- BioChemInsight: An Online Platform for Automated Extraction of Chemical Structures and Activity Data from Patents12 Apr
Alternatives
other vendors · similar capability- 10.2
- 9.7
- 8.5
- 8.4
Want GLM 4.5V on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.