Armand McQueen

basic-generation

ok

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

Produced 87 words (121 output tokens).

Result

{
  "wordCount": 87,
  "outputTokens": 121
}

e81e1885-8bd5-45f6-a160-0f6edee47063

Aug 8, 2026, 09:08:16

togetherai:moonshotai/Kimi-K3

Input

29

Output

121

Reasoning

0

Answer

121

Cost

$0.0019

Mode

streaming

Cost breakdown

Input $0.000087Output $0.0018

Latency

67.1s

TTFT

65.7s

Tok/s

1.8

Tok/s (adj)

82.9

usage-cost

ok

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

in=93 out=165 rt=148 cost=$0.002581 (paid).

Result

{
  "inputTokens": 93,
  "outputTokens": 165,
  "reasoningTokens": 148,
  "costUsd": 0.0025811999999999996,
  "paid": true
}

cc41346b-6d75-4fe5-888c-665e37639235

Aug 8, 2026, 09:08:35

togetherai:moonshotai/Kimi-K3

Input

93

Output

165

Reasoning

148

Answer

17

Cache Read

64

Cost

$0.0026

Mode

streaming

Cost breakdown

Input $0.000087Output $0.0025Cache read $0.000019

Latency

19.3s

TTFT

19.2s

Tok/s

8.5

Tok/s (adj)

230.3

streaming-structure

ok

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

mode=streaming, 23 stream events, 1 text block(s), 35 chars.

Result

{
  "mode": "streaming",
  "streamEventCount": 23,
  "textBlocks": 1,
  "textLen": 35
}

dc71e375-6173-4d5b-89a8-3e698de2c013

Aug 8, 2026, 09:08:50

togetherai:moonshotai/Kimi-K3

Input

98

Output

73

Reasoning

32

Answer

41

Cache Read

96

Cost

$0.0011

Mode

streaming

Cost breakdown

Input $0.000006Output $0.0011Cache read $0.000029

Latency

15.1s

TTFT

14.9s

Tok/s

4.8

Tok/s (adj)

190.6

complete-stream-parity

ok

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

stream="Paris" · complete="Paris"

Result

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

073406e1-7a2f-4d42-9290-0075a7055a59

Aug 8, 2026, 09:09:20

togetherai:moonshotai/Kimi-K3

Input

32

Output

7

Reasoning

0

Answer

7

Cost

$0.000201

Mode

streaming

Cost breakdown

Input $0.000096Output $0.000105

Latency

29.7s

TTFT

29.7s

Tok/s

0.2

Tok/s (adj)

161.5

64c9f0d7-1d0c-4760-b74c-b9b86b288abb

Aug 8, 2026, 09:09:48

togetherai:moonshotai/Kimi-K3

Input

32

Output

7

Reasoning

0

Answer

7

Cost

$0.000201

Mode

non-streaming

Cost breakdown

Input $0.000096Output $0.000105

Latency

27.6s

Tok/s

0.3

reasoning-conformance

ok

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

Class (from reasoning-config): toggleable.
off: out=332 · mid: out=418 (rt=181).

Result

{
  "class": "toggleable",
  "mode": "deterministic",
  "offOutputTokens": 332,
  "midOutputTokens": 418,
  "midReasoningTokens": 181
}

afc52a43-8357-4e75-acb4-ff5a69251a50

Aug 8, 2026, 09:09:56

togetherai:moonshotai/Kimi-K3

Input

103

Output

332

Reasoning

0

Answer

332

Cost

$0.0053

Mode

streaming

Cost breakdown

Input $0.000309Output $0.0050

Latency

7.9s

TTFT

2.1s

Tok/s

42.1

Tok/s (adj)

57.9

0b4fcb6e-02fe-4181-bb0b-1cba49811807

Aug 8, 2026, 09:10:03

togetherai:moonshotai/Kimi-K3

Input

103

Output

418

Reasoning

181

Answer

237

Cache Read

64

Cost

$0.0064

Mode

streaming

Cost breakdown

Input $0.000117Output $0.0063Cache read $0.000019

Latency

7.7s

TTFT

5.1s

Tok/s

54.0

Tok/s (adj)

90.2

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
}

faa12990-e594-480b-909a-b7df1ef5b20d

Aug 8, 2026, 09:10:07

togetherai:moonshotai/Kimi-K3

Input

172

Output

56

Reasoning

0

Answer

56

Cost

$0.0014

Mode

streaming

Cost breakdown

Input $0.000516Output $0.000840

Latency

3.7s

Tok/s

15.1

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: "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
}

b604eedb-bf94-4473-99b7-c5dbd283567a

Aug 8, 2026, 09:10:09

togetherai:moonshotai/Kimi-K3

Input

275

Output

50

Reasoning

0

Answer

50

Cache Read

256

Cost

$0.000884

Mode

streaming

Cost breakdown

Input $0.000057Output $0.000750Cache read $0.000077

Latency

1.9s

Tok/s

26.1

ee05d866-dcfd-4cdb-b0ae-8ac81f6992c5

Aug 8, 2026, 09:10:12

togetherai:moonshotai/Kimi-K3

Input

336

Output

35

Reasoning

14

Answer

21

Cache Read

192

Cost

$0.0010

Mode

streaming

Cost breakdown

Input $0.000432Output $0.000525Cache read $0.000058

Latency

2.5s

TTFT

2.4s

Tok/s

13.9

Tok/s (adj)

265.9

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.

reasoned=true, none token —; follow-up → "**Balance for account `acme-42`:** $1,234.00 **Growth calculation (5%"

Result

{
  "signatureType": "none",
  "tokenCaptured": false,
  "reasoned": true,
  "answered": true,
  "continuation": "**Balance for account `acme-42`:** $1,234.00\n\n**Growth calculation (5% annually, compounded for 7 years):**\n\n$$\\text{Future Value} = \\$1,234"
}

28c6864b-782d-4c7f-8c91-09dd17fadaaf

Aug 8, 2026, 09:10:19

togetherai:moonshotai/Kimi-K3

Input

205

Output

85

Reasoning

33

Answer

52

Cost

$0.0019

Mode

streaming

Cost breakdown

Input $0.000615Output $0.0013

Latency

7.0s

Tok/s

12.1

8af56bf9-73cf-486d-bd7e-39c5fe150b76

Aug 8, 2026, 09:10:33

togetherai:moonshotai/Kimi-K3

Input

289

Output

165

Reasoning

0

Answer

165

Cache Read

192

Cost

$0.0028

Mode

streaming

Cost breakdown

Input $0.000291Output $0.0025Cache read $0.000058

Latency

14.8s

TTFT

11.5s

Tok/s

11.1

Tok/s (adj)

49.8

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, United States"
  },
  "validJson": true,
  "missingKeys": [],
  "bareJson": true
}

3dd9bf5e-6156-4858-9385-49a485cfbf45

Aug 8, 2026, 09:10:36

togetherai:moonshotai/Kimi-K3

Input

87

Output

30

Reasoning

0

Answer

30

Cache Read

64

Cost

$0.000538

Mode

streaming

Cost breakdown

Input $0.000069Output $0.000450Cache read $0.000019

Latency

2.7s

TTFT

2.4s

Tok/s

11.2

Tok/s (adj)

95.6

summarize-memex

ok

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

Read ~10274 input tokens, wrote 457 words (627 output tokens) in 11785ms.

Result

{
  "wordCount": 457,
  "inputTokens": 10274,
  "outputTokens": 627,
  "latencyMs": 11785
}

8413d526-f92f-4232-a611-c795a0c4adfe

Aug 8, 2026, 09:10:48

togetherai:moonshotai/Kimi-K3

Input

10,274

Output

627

Reasoning

0

Answer

627

Cache Read

10,240

Cost

$0.0126

Mode

streaming

Cost breakdown

Input $0.000102Output $0.0094Cache read $0.0031

Latency

11.8s

TTFT

1.6s

Tok/s

53.2

Tok/s (adj)

61.3

truncation

ok

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

finishReason=length, out=100 tokens.

Result

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

efbcd000-d3a4-4bdd-920a-4e4c6fa15d8c

Aug 8, 2026, 09:11:19

togetherai:moonshotai/Kimi-K3

Input

35

Output

100

Reasoning

0

Answer

100

Cost

$0.0016

Mode

streaming

Cost breakdown

Input $0.000105Output $0.0015

Latency

31.4s

TTFT

27.8s

Tok/s

3.2

Tok/s (adj)

27.6

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"
}