Zhipu AI / manual · models.dev

GLM-5.2

Open flagship GLM for long-horizon coding agents and million-token context work

Encoding fertility 1.20×

Harnesses

Same model, different drivers. ChatGPT is not OpenCode. $ / MU is only set when every official task passed. $ / pass still shows on a partial run; it does not rank. Time is output tokens at 80 tok/s, not wall clock.

StackPassedAttemptsTime$ / MU$ / pass$ / hrTokens / passBurn$ / Zhipu AI
Qwenmax3/3632m 28s$0.000595$0.8521$4.74140,7021.00×$0.8521

Official jobs

Pass or fail on each sealed job. Attempts include retries. Failed tries stay in the bill.

Durable work queueJSON Patch (RFC 6902 subset)Token bucket and sliding window
pass · 4pass · 1pass · 1

Tokens by job

Billed tokens on each official job, including retries. Cache reads drop out of this count.

  • Qwen · max

    306,247Durable work queue
    56,360JSON Patch (RFC 6902 subset)
    59,500Token bucket and sliding window

Where the bill goes

Failed attempts stay in the bill. Cache hits are a discount, shown separately.

  • Qwen · max422,107

    Cache hit 5,760,896 (discount, not extra work)

  • Uncached in
  • Output
  • Thinking
  • Cache write

Retry tax

Attempts are harness calls, including failed hidden-verifier retries. Extra tries stay in the bill.

  • Qwenmax6 attempts · 3 extra / 3 jobs

Cache hits

Hits divided by uncached input plus hits. No cache means the harness did not report a discount.

  • Qwenmax96%

Encoding by slice

Native tokens per MU on the same frozen basket. Tool schemas are where fertility often doubles.

  • English prose30% of the basket0.93×
  • Code25% of the basket1.00×
  • Structured15% of the basket0.84×
  • Tool schemas10% of the basket1.03×
  • CJK10% of the basket3.48×
  • Instructions10% of the basket0.89×
SliceCharactersNative tokensFertilityTrue inSource
English prose4,4741,0400.93×$1.30estimate
Code2,3215831.00×$1.41estimate
Structured3,0676460.84×$1.18estimate
Tool schemas2,3786121.03×$1.44estimate
CJK8317233.48×$4.87estimate
Instructions1,1642600.89×$1.25estimate

Endpoints

ProviderSKUList inList outCache hitCache write
Zhipu AI
glm-5.2$1.40$4.40$0.26$0.00

All stacks

GLM-5.2 / Metered