BenchLeader

GLM 4.6V

Reasoning effort
Price#76 of 254
$0.450per 1M tokens
Cheaper than most
0.5× the median
Speed#199 of 218
34tokens/s
Bottom quarter
0.4× the median

No cheaper ranked model scores higher: it is on the price frontier.

As of 12 Nov 2026, GLM 4.6V by Zhipu AI is not yet ranked on the BenchLeader Index, which needs at least five independent results across two categories. API pricing: $0.300 per 1M input tokens and $0.900 per 1M output tokens ($0.450 blended at 3:1), list price on 12 Nov 2026. It streams about 34 tokens/s (Artificial Analysis). Released 8 Dec 2025.

Key facts (as of 12 Nov 2026)

  • GLM 4.6V API pricing: $0.300 per 1M input tokens and $0.900 per 1M output tokens ($0.450 blended at 3:1) (list price, 12 Nov 2026).
  • GLM 4.6V streams about 34 tokens/s (Artificial Analysis, 12 Nov 2026).
  • GLM 4.6V was released on 8 Dec 2025 (Zhipu AI).

In detail. GLM 4.6V is a Zhipu AI open-weights reasoning model, released 8 Dec 2025. It is not yet ranked: the index needs at least five independent results across two categories. At $0.450 per million tokens blended it is cheaper than most ranked models. Output speed of 34 tokens per second puts it in the slowest quarter, with a first answer in 7.6 s. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 8 Dec 2025.stale: no new result in six months

Blended price
$0.450/M
$0.300 in · $0.900 out
via Artificial Analysis
Output speed
34 tok/s
measured by Artificial Analysis
First answer
7.59 s
first token 7.59 s
Context
128k
via models.dev
Full answer
46 s
median, reasoning included
Cost per task
–
Tokens per task
–
Released
8 Dec 2025
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Agents & tools36
  2. Instruction following32
  3. Multimodal24
  4. Long context32

Get an email when GLM 4.6V gets its first benchmark results

And when Zhipu AI ships a new model, plus the weekly review on Mondays.

Choose providers and emails

Reasoning-effort configurations

The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.

EffortIndexRankSpeedFirst answerChat reply costCategories
no reasoningbest––34 tok/s7.59 s$0.0004Agents & tools 36 · Instruction following 32 · Long context 32 · Multimodal 24
thinking––39 tok/s29 s$0.0004Agents & tools 46 · Instruction following 34 · Long context 48 · Multimodal 30
not stated––36 tok/s3.38 s$0.0004Coding 54 · Human preference 55 · Multimodal 47 · Reasoning 53

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmarkno reasoningthinkingnot statedSourceTrend
Arena Hard Promptseffort not stated––1384#192Arena (LMArena)
GPQA Diamond (AA)not in index56.6%#40671.9%#288–Artificial Analysis
Humanity's Last Exam (AA)not in index3.7%#5989.6%#350–Artificial Analysis
CritPtnot in index0.0%#3350.0%#335–Artificial Analysis

Coding

Benchmarkno reasoningthinkingnot statedSourceTrend
Arena Codingeffort not stated––1416#191Arena (LMArena)
LiveCodeBench (AA)not in index41.1%#16116.0%#255–Artificial Analysis

Agents & tools

Benchmarkno reasoningthinkingSourceTrend
Terminal-Bench Hard3.0%#31914.4%#201Artificial Analysis
τ²-Bench Telecom (AA)not in index30.7%#26031.6%#256Artificial Analysis

Instruction following

Benchmarkno reasoningthinkingnot statedSourceTrend
IFBench27.9%#38130.1%#369–Artificial Analysis
Arena Instruction Followingnot in indexeffort not stated––1366#183Arena (LMArena)

Human preference

Benchmarknot statedSourceTrend
Arena Text1378#184Arena (LMArena)
Arena Creative Writingnot in index1349#174Arena (LMArena)
Arena Multi-turnnot in index1367#190Arena (LMArena)
Arena Longer Queriesnot in index1381#184Arena (LMArena)

Multimodal

Benchmarkno reasoningthinkingnot statedSourceTrend
Arena Visioneffort not stated––1163#111Arena (LMArena)
MMMU-Pro42.2%#27548.5%#257–Artificial Analysis

Long context

Benchmarkno reasoningthinkingSourceTrend
AA-LCR v1.117.0%#46848.7%#347Artificial Analysis

Composite

Benchmarkno reasoningthinkingSourceTrend
AA Intelligence Index v4.3.2not in index8.3#45011.2#367Artificial Analysis
AA Progress Index v1not in index177.1#484197.9#426Artificial Analysis

API pricing: where to run it

Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.

ProviderSpeedFirst tokenInput $/MOutput $/MContextQuantisation
NovitaAI34 tok/s4.61 s$0.300$0.900131kbf16
Z.ai32 tok/s0.81 s$0.300$0.900131kfp8

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.050 per 1M. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.0004$0.000316.4 s
Summarize a 30-page report12,000 / 600$0.0041$0.001925.1 s
Code edit6,000 / 1,500$0.0032$0.002051.4 s
Agentic coding session60,000 / 4,000$0.022$0.0102.1 min
Structured extraction2,000 / 200$0.0008$0.000413.4 s

More from Zhipu AI

New this month

See also

Questions people ask

Why does GLM 4.6V have no BenchLeader Index score?
The index needs at least five independent results across at least two categories, and as of 12 Nov 2026 GLM 4.6V does not have them yet. Launch-post scores are never used, so it is ranked once independent publishers have measured it.
How much does GLM 4.6V cost per 1M tokens?
As of 12 Nov 2026, GLM 4.6V costs $0.300 per 1M input tokens and $0.900 per 1M output tokens, $0.450 blended (three input tokens per output token).
When was GLM 4.6V released?
Zhipu AI released GLM 4.6V on 8 Dec 2025. BenchLeader has results for it at 3 reasoning-effort settings.

Data as of 12 Nov 2026. Compare these configurations.

Cite as: BenchLeader, “GLM 4.6V: benchmarks, pricing, speed and rank”, data as of 12 Nov 2026, https://www.benchleader.com/models/glm-4-6v.