Tier A — Frontier
Runs in:CNMade in:China
Z.ai (GLM / Zhipu)

GLM-4.7

Tier A — Frontier · 205K tokens

Tokonomix Editorial Team·Reviewed by Mes Kalkan··

GLM-4.7 is the most recent model in Zhipu’s well-established GLM-4 line, positioned as a strong general-purpose and coding/agentic model at a mid-market price. For many workloads it is the sweet spot of the GLM family — cheaper than the GLM-5 generation but still capable.

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 latency104 runs
1447896016474239873150008-1509-10ms
Section 02

Quality scores

How this model compares to the rest of the field on each prompt category, from a pairwise fit over the same prompts. The raw judge score sits underneath each number.

60%
Coding
judge mean 91
85%
Creative
judge mean 96
75%
Factual
judge mean 88
74%
Reasoning
judge mean 99

Win rate per category: how often this model beats a field-average model on a prompt from that category. 50% is average, not a failing grade. It is not a percentage of correct answers.

Section 03

Pricing history

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

💰
API rates — GLM-4.7
$0.6000 per 1M input tokens
$2.20 per 1M output tokens
≈ $0.0008 per typical conversation (800 tokens)
Input vs output price (per 1M tokens)
per 1M input tokens$0.6000
per 1M output tokens$2.20

Pricing over time

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

$0.6000

input / 1M

— stable

$2.20

output / 1M

— stable

2026-07-122026-08-162026-09-06
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)22 / avg 53
1379

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.toolsreasoning
Section 06

Availability

Availability

No measurements yet

We haven't recorded enough API calls to show availability stats for this model. Data appears once the model starts receiving live traffic.

Section 07

Tokonomix benchmark verdicts

⚖️
Endorsed by 1 judge
Independent LLM judges evaluated this model on our weekly intelligence tests
claude-sonnet-4-593/100 · 33 runs
30 correct0 partial3 wrong91% accuracy
● 2026-09-06

GLM-4.7 quality drops 22 points with latency degradation across benchmarks

GLM-4.7 experienced significant performance degradation in the current benchmark window, with overall quality falling from a perfect 100.0 to 77.7 points. The most severe decline occurred in coding performance, dropping from 100 to 63 points, representing a substantial regression in technical task handling. Creative performance remains the model's strongest category at 92 points, down from a perfect score but still demonstrating relative competency. Latency deteriorated notably, with p50 response times increasing 80% from 18413ms to 33220ms, now exceeding 33 seconds for typical requests. The benchmark window included only 4 test runs, matching the previous period's sample size. Factual and reasoning categories were not evaluated in the current window, making it unclear whether these capabilities have similarly regressed. The combination of quality decline and increased latency suggests potential issues with model deployment, infrastructure changes, or alterations to the underlying model configuration. Users should expect slower responses and reduced coding capabilities compared to the previous benchmark period, though creative tasks may still yield acceptable results.

Quality

77.7

Latency p50

33,220 ms

Test runs

4

✗ Quality dropped 22.3 points✗ Coding performance fell to 63✗ Latency increased 80%✓ Creative tasks remain strong
Section 08

Full model profile

GLM-4.7: the value workhorse of the GLM-4 line

GLM-4.7 is the most recent model in Zhipu’s well-established GLM-4 line, positioned as a strong general-purpose and coding/agentic model at a mid-market price. For many workloads it is the sweet spot of the GLM family — cheaper than the GLM-5 generation but still capable.

z.ai publishes GLM-4.7 at $0.60 per 1M input tokens and $2.20 per 1M output tokens — a mid-market price well below the GLM-5 generation.

It advertises a large context window (~200K tokens on our registration), suited to long files and multi-step agent runs.

Architecture & training signals

GLM-4.7 continues Zhipu AI’s GLM-4 line, which has been notable for capable open-weight releases with a focus on coding and agentic tool use. It is a reasoning-capable chat model with tool-calling and structured JSON over an OpenAI-compatible endpoint. 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

  • Coding and agentic tool-use — a traditional strength of the GLM-4 line.
  • Strong quality-per-euro; a good default when GLM-5 pricing is overkill.
  • Long-context tasks within its ~200K window.

Where it falls short

  • Not positioned as an absolute frontier model — for the hardest reasoning, GLM-5.2 or a US frontier model may edge it.
  • Independent Tokonomix benchmark data is not in yet; verify on your tasks.
  • Non-EU hosting.

Real-world use cases

  • Coding assistants, refactors and agentic build loops.
  • General drafting, analysis and summarisation at scale.
  • A cost-effective, decorrelated consensus proposer.

Tokonomix benchmark snapshot

GLM-4.7 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-4.7 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-4.7 is the GLM family’s value workhorse: capable, coding-friendly and priced for volume. Reach for GLM-5.x only when you specifically need the newest generation’s reasoning; for zero-cost work, the GLM-4.7 Flash free tier is the natural companion.

Last automated test
Sep 10, 2026 · 08:03 UTC · Speed benchmark
P50 latency
8989 ms
P95 latency
22941 ms
Errors
0 / 6 runs
Last reviewed by Tokonomix Team·July 8, 2026