together:glm-5.3

9/5/2026, 2:28:11 AM · 15 tests · run 9f5040b1-4c2a-48a9-85cb-dfbd5b439fb2

basic-generation

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

Produced 3 words (286 output tokens).

Result

{
  "wordCount": 3,
  "outputTokens": 286
}

e0a8533e-8c18-40d8-9dfa-f4e43c1226f3

Sep 5, 2026, 02:21:10

togetherai:zai-org/GLM-5.3

Input

Output

286

Reasoning

202

Answer

Cost

$0.0013

Mode

streaming

Cost breakdown

Input $0.000032Output $0.0013

Latency

7.2s

TTFT

6.6s

Tok/s

39.8

Tok/s (adj)

145.9

usage-cost

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

in=20 out=116 rt=109 cost=$0.000538 (paid).

Result

{
  "inputTokens": 20,
  "outputTokens": 116,
  "reasoningTokens": 109,
  "costUsd": 0.0005384000000000001,
  "paid": true
}

0960e9e1-108d-49cb-b4a2-bcf3dd808482

Sep 5, 2026, 02:25:19

togetherai:zai-org/GLM-5.3

Input

Output

116

Reasoning

109

Answer

Cost

$0.000538

Mode

streaming

Cost breakdown

Input $0.000028Output $0.000510

Latency

249.1s

TTFT

249.1s

Tok/s

0.5

Tok/s (adj)

253.7

streaming-structure

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

mode=streaming, 24 stream events, 1 text block(s), 35 chars.

Result

{
  "mode": "streaming",
  "streamEventCount": 24,
  "textBlocks": 1,
  "textLen": 35
}

5c2c8d52-bd20-4f0d-8c03-7c29856777e4

Sep 5, 2026, 02:25:20

togetherai:zai-org/GLM-5.3

Input

Output

Reasoning

Answer

Cost

$0.000361

Mode

streaming

Cost breakdown

Input $0.000035Output $0.000326

Latency

547ms

TTFT

471ms

Tok/s

135.3

Tok/s (adj)

408.8

complete-stream-parity

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

stream="Paris" · complete="Paris"

Result

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

3a0d3c25-b913-4477-b11f-9057f19c9fbb

Sep 5, 2026, 02:25:20

togetherai:zai-org/GLM-5.3

Input

Output

Reasoning

Answer

Cost

$0.000050

Mode

streaming

Cost breakdown

Input $0.000036Output $0.000013

Latency

243ms

TTFT

238ms

Tok/s

12.3

Tok/s (adj)

619.8

8c2e6c66-fe6b-41dd-b70a-90e6ac5da816

Sep 5, 2026, 02:25:20

togetherai:zai-org/GLM-5.3

Input

Output

Reasoning

Answer

Cost

$0.000050

Mode

non-streaming

Cost breakdown

Input $0.000036Output $0.000013

Latency

298ms

Tok/s

10.1

reasoning-conformance

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

Class (from reasoning-config): always-on.
reasoningTokens=95, outputTokens=414.

Result

{
  "class": "always-on",
  "reasoningTokens": 95,
  "outputTokens": 414
}

fb41fb64-55b1-4dc3-a97e-6877a1a1c3c9

Sep 5, 2026, 02:25:22

togetherai:zai-org/GLM-5.3

Input

Output

414

Reasoning

Answer

319

Cost

$0.0020

Mode

streaming

Cost breakdown

Input $0.000136Output $0.0018

Latency

1.2s

TTFT

403ms

Tok/s

339.3

Tok/s (adj)

390.6

single-tool-call

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
}

9da6b399-fa07-47d8-ab3d-13396dbb3c93

Sep 5, 2026, 02:25:22

togetherai:zai-org/GLM-5.3

Input

230

Output

Reasoning

Answer

Cost

$0.000397

Mode

streaming

Cost breakdown

Input $0.000322Output $0.000075

Latency

212ms

Tok/s

80.2

tool-loop

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

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

Result

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

f17b3e2d-ce7e-4eb8-93f3-e93b1cfbd13a

Sep 5, 2026, 02:26:45

togetherai:zai-org/GLM-5.3

Input

226

Output

Reasoning

Answer

Cost

$0.000426

Mode

streaming

Cost breakdown

Input $0.000316Output $0.000110

Latency

83.2s

Tok/s

0.3

f50d8001-efd8-4024-873b-dbb98ce42d80

Sep 5, 2026, 02:27:27

togetherai:zai-org/GLM-5.3

Input

253

Output

Reasoning

Answer

Cache Read

192

Cost

$0.000486

Mode

streaming

Cost breakdown

Input $0.000085Output $0.000132Cache read $0.000269

Latency

42.0s

TTFT

41.2s

Tok/s

0.7

Tok/s (adj)

15.9

web-search

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

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

reasoned=true, none token —; follow-up → "**Current balance:** $1,234 (account acme-42) **Calculation for 7 yea"

Result

{
  "signatureType": "none",
  "tokenCaptured": false,
  "reasoned": true,
  "answered": true,
  "continuation": "**Current balance:** $1,234 (account acme-42)\n\n**Calculation for 7 years at 5% compounded annually:**\n\nBalance = 1234 × (1.05)⁷\n\n- 1.05⁷ = 1"
}

ca1936ed-a3a3-4e34-8fe8-2cb75e7cd5af

Sep 5, 2026, 02:27:49

togetherai:zai-org/GLM-5.3

Input

253

Output

Reasoning

Answer

Cost

$0.000455

Mode

streaming

Cost breakdown

Input $0.000354Output $0.000101

Latency

22.2s

Tok/s

1.0

b83147fb-5a20-4950-91c3-22ee0b8dd856

Sep 5, 2026, 02:27:54

togetherai:zai-org/GLM-5.3

Input

288

Output

384

Reasoning

177

Answer

207

Cost

$0.0021

Mode

streaming

Cost breakdown

Input $0.000403Output $0.0017

Latency

4.5s

TTFT

2.5s

Tok/s

85.5

Tok/s (adj)

103.6

prompt-cache

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

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, USA"
  },
  "validJson": true,
  "missingKeys": [],
  "bareJson": true
}

05f7bcc9-45c1-4476-875e-76358a155f47

Sep 5, 2026, 02:27:54

togetherai:zai-org/GLM-5.3

Input

Output

Reasoning

Answer

Cost

$0.000226

Mode

streaming

Cost breakdown

Input $0.000098Output $0.000128

Latency

400ms

TTFT

305ms

Tok/s

72.5

Tok/s (adj)

304.5

summarize-memex

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

Read ~10238 input tokens, wrote 713 words (911 output tokens) in 12756ms.

Result

{
  "wordCount": 713,
  "inputTokens": 10238,
  "outputTokens": 911,
  "latencyMs": 12756
}

ffab58e7-e9b1-495f-a3c8-5085f1f168cf

Sep 5, 2026, 02:28:07

togetherai:zai-org/GLM-5.3

Input

10,238

Output

911

Reasoning

Answer

911

Cost

$0.0183

Mode

streaming

Cost breakdown

Input $0.0143Output $0.0040

Latency

12.8s

TTFT

1.3s

Tok/s

71.4

Tok/s (adj)

79.2

truncation

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

finishReason=length, out=100 tokens.

Result

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

5db0cb8a-b754-4095-a11b-c52566291355

Sep 5, 2026, 02:28:08

togetherai:zai-org/GLM-5.3

Input

Output

100

Reasoning

Answer

100

Cost

$0.000481

Mode

streaming

Cost breakdown

Input $0.000041Output $0.000440

Latency

1.4s

TTFT

265ms

Tok/s

69.0

Tok/s (adj)

84.4

streaming-only-rejection

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

streamingOnly=false; complete() succeeded.

Result

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

abort

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

pre-aborted signal → rejected (AbortError).

Result

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