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

GLM-4.6

Tier A — Frontier · 205K tokens

Tokonomix Editorial Team·Reviewed by Mes Kalkan··

GLM-4.6 is an established model in Zhipu’s GLM-4 line, built around a large context window and a focus on coding and agentic tool use. It offers most of GLM-4.7’s practical value and is one of the more mature GLM chat models.

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 latency56 runs
1004862816252238763150007-0807-22ms
Section 02

Quality scores

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

93
Coding
0
Creative
Section 03

Pricing history

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

💰
API rates — GLM-4.6
$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

2026-07-122026-07-192026-07-19
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)29 / avg 61
19812

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

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=17

Median response time

65,035ms

n=17

Based on 191 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)

17

OK responses (30d)

17

Total calls (7d)

0

OK responses (7d)

0

Section 07

Tokonomix benchmark verdicts

⚖️
Endorsed by 1 judge
Independent LLM judges evaluated this model on our weekly intelligence tests
claude-sonnet-4-577/100 · 5 runs
4 correct0 partial1 wrong80% accuracy
● 2026-07-19

GLM-4.6 maintains strong coding with new JSON and tool capabilities

GLM-4.6 from Zhipu continues to demonstrate solid performance in the current benchmark window, building on its established strengths. The model shows consistent coding capabilities, maintaining its position as a competent option for development tasks. New capabilities have been added including native JSON output formatting, tool use functionality, and reasoning features, expanding the model's practical applications beyond its core competencies. Performance remains stable across key evaluation areas, with the model showing particular reliability in structured tasks. The addition of JSON mode and tool calling capabilities makes GLM-4.6 more versatile for integration into production workflows where formatted outputs and function calling are essential. The reasoning capability addition suggests an attempt to handle more complex multi-step problems, though the extent of improvement in this area would require further evaluation. For developers and teams considering GLM-4.6, the model presents a balanced option with established coding performance now supplemented by practical features like tool use and structured output. The stability in core performance combined with incremental capability additions positions it as a reliable choice for applications requiring consistent behavior alongside modern API features.

Quality

—

Latency p50

—

Test runs

0

✓ JSON output mode added✓ Tool calling capability enabled✓ Reasoning feature introduced✓ Stable coding performance maintained
Section 08

Full model profile

GLM-4.6: long-context, coding-focused GLM

GLM-4.6 is an established model in Zhipu’s GLM-4 line, built around a large context window and a focus on coding and agentic tool use. It offers most of GLM-4.7’s practical value and is one of the more mature GLM chat models.

z.ai publishes GLM-4.6 at $0.60 per 1M input tokens and $2.20 per 1M output tokens, the same value tier as GLM-4.7.

It advertises a large ~200K-token context window — a defining feature of GLM-4.6 for long-file and repository-scale work.

Architecture & training signals

GLM-4.6 is part of the GLM-4 line from Zhipu AI, which earned attention for capable open-weight models tuned for coding and agentic tasks and for long context. It is a reasoning-capable chat model with tool-calling and 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

  • Long-context work — its large window is the headline feature.
  • Coding and agentic tool use.
  • A cost-effective, mature GLM option.

Where it falls short

  • GLM-4.7 and the GLM-5 generation may edge it on newer capabilities.
  • No Tokonomix benchmark scores yet; validate on your workload.
  • Non-EU hosting.

Real-world use cases

  • Repository-scale code understanding and generation.
  • Long-document analysis where the full text must fit in one call.
  • Agentic pipelines needing reliable tool-calls.

Tokonomix benchmark snapshot

GLM-4.6 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-4.6 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-4.6 is a dependable, long-context, coding-focused GLM at a value price. If you want the newest iteration, GLM-4.7; if you want the newest generation entirely, GLM-5.x. For free experimentation, pair it with the GLM Flash tiers.

Last automated test
Jul 22, 2026 · 02:00 UTC · Speed benchmark
P50 latency
6823 ms
P95 latency
7640 ms
Errors
0 / 6 runs
Last reviewed by Tokonomix Team·July 8, 2026