Armand McQueen

basic-generation

ok

Plain streaming generation with a soft length constraint (~100 words).

Produced 90 words (110 output tokens).

Result

{
  "wordCount": 90,
  "outputTokens": 110
}

4f541e2e-4449-4734-b5c2-c92ba8a35839

Aug 17, 2026, 07:10:02

deepinfra:zai-org/GLM-5.2

Input

17

Output

110

Reasoning

0

Answer

110

Cost

$0.000277

Mode

streaming

Cost breakdown

Input $0.000013Output $0.000264

Latency

16.0s

TTFT

950ms

Tok/s

6.9

Tok/s (adj)

7.3

usage-cost

ok

Asserts usage tokens populate and cost computes from the ChatReq contract.

in=20 out=174 rt=0 cost=$0.000433 (paid).

Result

{
  "inputTokens": 20,
  "outputTokens": 174,
  "reasoningTokens": 0,
  "costUsd": 0.0004326,
  "paid": true
}

b718fcaf-7ab2-4c5b-b963-688a45a1a59a

Aug 17, 2026, 07:10:06

deepinfra:zai-org/GLM-5.2

Input

20

Output

174

Reasoning

0

Answer

174

Cost

$0.000433

Mode

streaming

Cost breakdown

Input $0.000015Output $0.000418

Latency

3.5s

TTFT

3.4s

Tok/s

50.3

Tok/s (adj)

4926.4

streaming-structure

ok

Asserts a streaming call captures events that reconstruct into the final text.

mode=streaming, 70 stream events, 1 text block(s), 35 chars.

Result

{
  "mode": "streaming",
  "streamEventCount": 70,
  "textBlocks": 1,
  "textLen": 35
}

b9906493-b866-47c8-86f9-f3a6fb80f2ae

Aug 17, 2026, 07:10:21

deepinfra:zai-org/GLM-5.2

Input

25

Output

227

Reasoning

0

Answer

227

Cost

$0.000564

Mode

streaming

Cost breakdown

Input $0.000019Output $0.000545

Latency

15.3s

TTFT

12.6s

Tok/s

14.8

Tok/s (adj)

82.7

complete-stream-parity

ok

Same prompt via stream() and complete() both return valid, matching results.

stream="Paris" · complete="Paris"

Result

{
  "streamText": "Paris",
  "completeText": "Paris"
}

9de26a78-65b0-4ae5-944f-e4fe7e0a2d94

Aug 17, 2026, 07:10:25

deepinfra:zai-org/GLM-5.2

Input

20

Output

2

Reasoning

0

Answer

2

Cost

$0.000020

Mode

streaming

Cost breakdown

Input $0.000015Output $0.000005

Latency

4.4s

TTFT

4.2s

Tok/s

0.5

Tok/s (adj)

9.5

594a35f9-e7ff-4774-88a0-4aa0c139128e

Aug 17, 2026, 07:10:26

deepinfra:zai-org/GLM-5.2

Input

20

Output

2

Reasoning

0

Answer

2

Cost

$0.000020

Mode

non-streaming

Cost breakdown

Input $0.000015Output $0.000005

Latency

679ms

Tok/s

2.9

reasoning-conformance

ok

Asserts the model's reasoning knob behaves as its reasoning-config declares.

Class (from reasoning-config): toggleable.
off: out=542 · mid: out=1222 (rt=0).

Result

{
  "class": "toggleable",
  "mode": "deterministic",
  "offOutputTokens": 542,
  "midOutputTokens": 1222,
  "midReasoningTokens": 0
}

9f2b55b5-23d0-4b80-8966-0e7cbe1bbbe7

Aug 17, 2026, 07:10:40

deepinfra:zai-org/GLM-5.2

Input

91

Output

542

Reasoning

0

Answer

542

Cost

$0.0014

Mode

streaming

Cost breakdown

Input $0.000068Output $0.0013

Latency

14.0s

TTFT

305ms

Tok/s

38.6

Tok/s (adj)

39.5

099f5053-a732-4b02-8056-a5ae2c0a79ff

Aug 17, 2026, 07:11:02

deepinfra:zai-org/GLM-5.2

Input

97

Output

1,222

Reasoning

0

Answer

1,222

Cost

$0.0030

Mode

streaming

Cost breakdown

Input $0.000073Output $0.0029

Latency

21.7s

TTFT

16.1s

Tok/s

56.2

Tok/s (adj)

218.1

single-tool-call

ok

Offers a calculator tool and checks the model calls it for an arithmetic question.

Called "multiply" with args {"a":23,"b":17}.

Result

{
  "toolCalled": true,
  "toolName": "multiply",
  "args": {
    "a": 23,
    "b": 17
  },
  "correct": true
}

c8b568a4-d2d8-4299-8007-45b6c42c20b4

Aug 17, 2026, 07:11:03

deepinfra:zai-org/GLM-5.2

Input

224

Output

16

Reasoning

0

Answer

16

Cost

$0.000206

Mode

streaming

Cost breakdown

Input $0.000168Output $0.000038

Latency

950ms

Tok/s

16.8

tool-loop

ok

Forces a tool call, feeds the result back, and checks the continuation uses it.

step1 (required): multiply({"a":23,"b":17}) → step2: "The product of 23 multiplied by 17 is **391**."

Result

{
  "toolName": "multiply",
  "args": {
    "a": 23,
    "b": 17
  },
  "product": 391,
  "continuation": "The product of 23 multiplied by 17 is **391**.",
  "usesResult": true,
  "forced": true
}

35c3d543-5f6b-49ac-9f99-c59710a407db

Aug 17, 2026, 07:11:28

deepinfra:zai-org/GLM-5.2

Input

226

Output

1,024

Reasoning

0

Answer

1,024

Cost

$0.0026

Mode

streaming

Cost breakdown

Input $0.000170Output $0.0025

Latency

24.9s

Tok/s

41.1

76fba2b2-2900-4f66-af7e-cd5f5f2f12ed

Aug 17, 2026, 07:11:30

deepinfra:zai-org/GLM-5.2

Input

1,798

Output

48

Reasoning

0

Answer

48

Cache Read

1,152

Cost

$0.000761

Mode

streaming

Cost breakdown

Input $0.000485Output $0.000115Cache read $0.000161

Latency

1.7s

TTFT

1.6s

Tok/s

28.0

Tok/s (adj)

365.4

web-search

ok

Enables server-side web search and checks invocation counts, the per-request cost surcharge, and provider-executed output blocks.

Skipped: model does not support the catalog's web_search tool version.

Result

{
  "skipped": "no server-side web search support"
}

reasoning-tool-continuity

ok

Reasoning model: continuity token captured on a tool call and replayed so the follow-up succeeds.

Step 1 didn't engage reasoning (adaptive discretion) — signature not exercised this run.
reasoned=false, none token —; follow-up → "Great, the current balance for account **acme-42** is **$1,234.00**. N"

Result

{
  "signatureType": "none",
  "tokenCaptured": false,
  "reasoned": false,
  "answered": true,
  "continuation": "Great, the current balance for account **acme-42** is **$1,234.00**. Now let me work through the compound-interest calculation.\n\n---\n\n### St"
}

80db2df5-2cd8-4b44-b893-a5df46300788

Aug 17, 2026, 07:11:36

deepinfra:zai-org/GLM-5.2

Input

253

Output

78

Reasoning

0

Answer

78

Cache Read

128

Cost

$0.000299

Mode

streaming

Cost breakdown

Input $0.000094Output $0.000187Cache read $0.000018

Latency

2.7s

Tok/s

29.4

4413cff7-ac9d-4094-ac6c-e6248fd56d98

Aug 17, 2026, 07:11:53

deepinfra:zai-org/GLM-5.2

Input

337

Output

497

Reasoning

0

Answer

497

Cost

$0.0014

Mode

streaming

Cost breakdown

Input $0.000253Output $0.0012

Latency

17.3s

TTFT

5.7s

Tok/s

28.7

Tok/s (adj)

42.8

prompt-cache

ok

Repeat a large prefix; assert cache-read tokens are reported and cost drops.

Skipped: provider has no (known) prompt-cache reporting.

Result

{
  "skipped": "cacheStyle=none"
}

json-output

ok

Requests a strict JSON object and checks it parses with the required keys.

Output parsed as JSON.
All required keys present.
Output is bare JSON (no leading prose/fence).

Result

{
  "parsed": {
    "name": "Anthropic",
    "founded": 2021,
    "headquarters": "San Francisco, California"
  },
  "validJson": true,
  "missingKeys": [],
  "bareJson": true
}

1880c878-0a64-48a7-94d0-0caf999d234a

Aug 17, 2026, 07:11:54

deepinfra:zai-org/GLM-5.2

Input

64

Output

26

Reasoning

0

Answer

26

Cost

$0.000110

Mode

streaming

Cost breakdown

Input $0.000048Output $0.000062

Latency

538ms

TTFT

410ms

Tok/s

48.3

Tok/s (adj)

202.6

summarize-memex

ok

Summarizes an ~11k-token essay to ~500 words (long input, reasoning minimized).

Read ~10232 input tokens, wrote 601 words (752 output tokens) in 19667ms.

Result

{
  "wordCount": 601,
  "inputTokens": 10232,
  "outputTokens": 752,
  "latencyMs": 19667
}

d8011e4b-e6d5-4cd6-92e6-a6dcc22d80a7

Aug 17, 2026, 07:12:14

deepinfra:zai-org/GLM-5.2

Input

10,232

Output

752

Reasoning

0

Answer

752

Cost

$0.0095

Mode

streaming

Cost breakdown

Input $0.0077Output $0.0018

Latency

19.7s

TTFT

1.7s

Tok/s

38.2

Tok/s (adj)

42.0

truncation

ok

Tiny maxTokens on a long prompt yields finishReason:"length".

finishReason=length, out=100 tokens.

Result

{
  "finishReason": "length",
  "outputTokens": 100
}

51aee1d4-971b-4d2b-8ee9-7a87fa38e9f5

Aug 17, 2026, 07:12:15

deepinfra:zai-org/GLM-5.2

Input

23

Output

100

Reasoning

0

Answer

100

Cost

$0.000257

Mode

streaming

Cost breakdown

Input $0.000017Output $0.000240

Latency

1.8s

TTFT

499ms

Tok/s

55.7

Tok/s (adj)

77.2

streaming-only-rejection

ok

Streaming-only models reject complete(); all others accept it.

streamingOnly=false; complete() succeeded.

Result

{
  "streamingOnly": false,
  "completeThrew": false
}

abort

ok

A pre-aborted signal rejects the call (abort is threaded + honored).

pre-aborted signal → rejected (AbortError).

Result

{
  "rejected": true,
  "errorName": "AbortError"
}