Tier A — Frontier

Runs in:CNMade in:China

GLM-4.5V (vision)

Tier A — Frontier · 66K tokens

Tokonomix Editorial Team·Reviewed by Mes Kalkan·Published July 6, 2026·Last reviewed July 8, 2026

GLM-4.5V is the vision-capable member of the GLM-4.5 line: it accepts images with text and reasons over both. It is a capable multimodal option anchored in the well-understood GLM-4.5 generation.

Section 01

Speed analysis

Latency measured across all benchmark runs. P50 (median) and P95 (95th percentile) give a realistic picture of response speed under normal and peak load.

P50 latency (median)P95 latency57 runs

Section 02

Quality scores

Evaluation results from judge-model scoring across diverse task categories. Scores reflect coherence, accuracy and instruction-following.

Coding

Creative

Section 03

Pricing history

Direct provider rates per million tokens, plus a typical-conversation cost estimate.

💰

API rates — GLM-4.5V (vision)

$0.6000 per 1M input tokens

$1.80 per 1M output tokens

≈ $0.0007 per typical conversation (800 tokens)

Input vs output price (per 1M tokens)

per 1M input tokens$0.6000

per 1M output tokens$1.80

Pricing over time

Input & output per 1M tokens · step-line = price changes

$0.6000

input / 1M

— stable

$1.80

output / 1M

— stable

2026-07-122026-07-192026-07-19

Input

Output

Price change

⟳ synced weekly

Section 04

Tokens per second

Throughput in tokens per second, derived from measured P50 latency. Higher is better; fluctuations track provider-side load.

Throughput (tokens / s)122 / avg 139

Estimated from P50 latency × 200 output tokens — the absolute number depends on this assumption; the trend is what matters.

Section 05

Capabilities

jsonnotes: GLM emits a non-standard reasoning_content field beside content; read content for the answer.toolsvision

Section 06

Availability

How often this model answers when we call it — measured across real API requests and live tests over the last 30 days. This is separate from quality: these numbers only tell you whether the model responds, not how good the answer is.

Last 7 days

—

Last 30 days

100.0%

n=1

Median response time

1,862ms

n=1

Based on 178 measurements over the last 30 days.

Technical details

Only live API calls and live-test requests count — internal probes and benchmark runs are excluded.

Calls with a custom API key (BYOK) are excluded: those failures are key-specific, not a sign of model downtime.

Failed calls are NOT included in quality scores — quality is measured on successful responses only. Availability and quality are independent signals.

Median response time (p50) across successful calls with a recorded duration. Outliers (very slow or very fast calls) pull the median less than the average.

Total calls (30d)

OK responses (30d)

Total calls (7d)

OK responses (7d)

Section 07

Tokonomix benchmark verdicts

⚖️

Endorsed by 1 judge

Independent LLM judges evaluated this model on our weekly intelligence tests

claude-sonnet-4-558/100 · 6 runs

2 correct1 partial3 wrong33% accuracy

● 2026-07-19

GLM-4.5V debuts with strong vision and multimodal reasoning capabilities

GLM-4.5V enters the benchmark landscape as Zhipu's new vision-capable model, demonstrating solid performance across multimodal tasks. The model shows particular strength in document understanding and visual reasoning, with competitive scores on common vision benchmarks. Its text-based performance aligns with mid-tier language models, offering reliable coding assistance and general instruction following. The addition of JSON mode and tool calling extends its utility for structured outputs and agent-based workflows, making it suitable for applications requiring both visual analysis and programmatic integration. While the model doesn't lead in any single category, it presents a well-balanced profile for developers seeking vision capabilities within the GLM ecosystem. Performance appears stable across different types of visual inputs, from natural images to technical diagrams. The model's multimodal reasoning shows promise for practical applications like document processing, image analysis, and visual question answering. Users should expect competent performance for most standard vision-language tasks, though cutting-edge applications may still require models with higher benchmark scores in specific domains.

Quality

—

Latency p50

—

Test runs

✓ Vision capabilities added✓ JSON and tool support✓ Strong multimodal reasoning

Section 08

Full model profile

GLM-4.5V: the GLM-4.5 vision model

GLM-4.5V is the vision-capable member of the GLM-4.5 line: it accepts images with text and reasons over both. It is a capable multimodal option anchored in the well-understood GLM-4.5 generation.

z.ai publishes GLM-4.5V at $0.60 per 1M input tokens and $1.80 per 1M output tokens.

It advertises a 64K-token context window — smaller than GLM-4.6V, but ample for typical image-plus-prompt tasks.

Architecture & training signals

GLM-4.5V is the vision variant of Zhipu AI’s GLM-4.5, adding image input to the text model. It produces text output reasoning over the supplied images, with tool and JSON support over an OpenAI-compatible endpoint. Output modality is text (image understanding, not generation). Like the rest of the GLM line, it returns a non-standard reasoning_content field alongside content in its OpenAI-compatible responses; integrations should read content for the final answer and treat reasoning_content as an optional trace.

Where it shines

Image understanding grounded in the mature GLM-4.5 generation.
Visual QA, extraction and description tasks.
Tool-calling and JSON for multimodal agents.

Where it falls short

Smaller context window than GLM-4.6V; and it reads, not generates, images.
No Tokonomix benchmark data yet; validate on your images.
Non-EU hosting.

Real-world use cases

Screenshot and document understanding.
Visual question answering.
Multimodal analysis pipelines.

Tokonomix benchmark snapshot

GLM-4.5V is newly registered on Tokonomix and not yet activated, so we have not run it through our weekly intelligence test or speed benchmark. There are no Tokonomix scores to report yet — and we will not invent any.

When it goes live, it enters the same weekly harness as every other model: identical prompts, an independent cross-family judge, and reproducible latency and cost measurements. Until then, treat the pricing and capability notes on this page as the vendor-published starting point, not as measured Tokonomix results.

EU privacy & data residency

GLM-4.5V is built by Zhipu AI (z.ai), a China-headquartered lab, and is served from non-EU infrastructure. This is important to state plainly: routing a prompt to this model is not an EU-data-residency or GDPR-sovereign choice, and Tokonomix will never tag it as one.

If your use case requires data to stay within the EU, pick a model whose provider is EU-hosted (for example our OVH or Azure-EU routes) rather than a GLM model. Tokonomix keeps z.ai out of every EU-only / sovereign routing set by design. Use GLM where its capability or price is the priority and cross-border processing is acceptable for that workload.

Verdict & alternatives

GLM-4.5V is a solid GLM-4.5-generation vision model. For a larger context window and a lower price, GLM-4.6V is the newer pick; for zero cost, GLM-4.6V Flash.

Last automated test

Jul 22, 2026 · 08:04 UTC · Speed benchmark

P50 latency

1633 ms

P95 latency

8229 ms

Errors

0 / 6 runs

Last reviewed by Tokonomix Team·July 8, 2026