Armand McQueen

basic-generation

ok

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

Produced 100 words (243 output tokens).

Result

{
  "wordCount": 100,
  "outputTokens": 243
}

3fecbd4e-9bc0-4ad6-858b-8cea8b5df2af

Jul 6, 2026, 21:55:48

anthropic:claude-fable-5

Input

20

Output

243

Reasoning

28

Answer

215

Cost

$0.0123

Mode

streaming

Cost breakdown

Input $0.000200Output $0.0121

Latency

7.1s

TTFT

4.0s

Tok/s

34.4

Tok/s (adj)

69.9

usage-cost

ok

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

in=20 out=11 rt=0 cost=$0.000750 (paid).

Result

{
  "inputTokens": 20,
  "outputTokens": 11,
  "reasoningTokens": 0,
  "costUsd": 0.00075,
  "paid": true
}

e61c9d05-2538-4bf4-a98d-9f52410c0d83

Jul 6, 2026, 21:55:52

anthropic:claude-fable-5

Input

20

Output

11

Reasoning

0

Answer

11

Cost

$0.000750

Mode

streaming

Cost breakdown

Input $0.000200Output $0.000550

Latency

3.4s

TTFT

2.5s

Tok/s

3.3

Tok/s (adj)

12.8

streaming-structure

ok

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

mode=streaming, 10 stream events, 1 text block(s), 35 chars.

Result

{
  "mode": "streaming",
  "streamEventCount": 10,
  "textBlocks": 1,
  "textLen": 35
}

62cd62d7-fc2d-4dd4-b147-13d3a9ae0f94

Jul 6, 2026, 21:55:56

anthropic:claude-fable-5

Input

21

Output

38

Reasoning

7

Answer

31

Cost

$0.0021

Mode

streaming

Cost breakdown

Input $0.000210Output $0.0019

Latency

4.1s

TTFT

4.1s

Tok/s

9.3

Tok/s (adj)

2403.1

complete-stream-parity

ok

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

stream="Paris" · complete="Paris"

Result

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

7d88a917-1615-42eb-89e7-5b531025f919

Jul 6, 2026, 21:55:58

anthropic:claude-fable-5

Input

23

Output

5

Reasoning

0

Answer

5

Cost

$0.000480

Mode

streaming

Cost breakdown

Input $0.000230Output $0.000250

Latency

2.8s

TTFT

2.8s

Tok/s

1.8

Tok/s (adj)

191.9

d7f33f93-db46-4584-a3cb-378a7f21b17e

Jul 6, 2026, 21:56:02

anthropic:claude-fable-5

Input

23

Output

5

Cost

$0.000480

Mode

non-streaming

Cost breakdown

Input $0.000230Output $0.000250

Latency

3.3s

Tok/s

1.5

reasoning-conformance

ok

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

Class (from reasoning-config): always-on.
reasoningTokens=122, outputTokens=621.

Result

{
  "class": "always-on",
  "reasoningTokens": 122,
  "outputTokens": 621
}

37a387c2-6a7c-45bd-802c-82b60d461ed6

Jul 6, 2026, 21:56:13

anthropic:claude-fable-5

Input

127

Output

621

Reasoning

122

Answer

499

Cost

$0.0323

Mode

streaming

Cost breakdown

Input $0.0013Output $0.0311

Latency

11.6s

TTFT

4.9s

Tok/s

53.6

Tok/s (adj)

74.8

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
}

a3bdaf25-3c76-4cc6-b746-9e28399ee71e

Jul 6, 2026, 21:56:17

anthropic:claude-fable-5

Input

472

Output

68

Reasoning

0

Answer

68

Cost

$0.0081

Mode

streaming

Cost breakdown

Input $0.0047Output $0.0034

Latency

3.8s

Tok/s

17.7

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
}

7b850669-6239-48df-b743-b114c4618887

Jul 6, 2026, 21:56:21

anthropic:claude-fable-5

Input

595

Output

58

Reasoning

0

Answer

58

Cost

$0.0089

Mode

streaming

Cost breakdown

Input $0.0060Output $0.0029

Latency

4.1s

Tok/s

14.2

b575363b-e6b1-4d68-af3e-57ff7572f0ca

Jul 6, 2026, 21:56:24

anthropic:claude-fable-5

Input

554

Output

21

Reasoning

0

Answer

21

Cost

$0.0066

Mode

streaming

Cost breakdown

Input $0.0055Output $0.0010

Latency

2.7s

TTFT

2.7s

Tok/s

7.6

Tok/s (adj)

1216.0

web-search

ok

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

Ran 1 web search(es); surcharge 0.0100 USD on top of token cost.

Result

{
  "searches": 1,
  "surchargeUsd": 0.01
}

955156fc-c3d7-41ee-b0b6-61ed476aaa09

Jul 6, 2026, 21:56:54

anthropic:claude-fable-5

Input

20,811

Output

1,059

Reasoning

212

Answer

847

Web searches

1

Cost

$0.2711

Mode

streaming

Cost breakdown

Input $0.2081Output $0.0530Web search $0.0100

Latency

30.4s

TTFT

30.3s

Tok/s

34.8

Tok/s (adj)

9316.9

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, anthropic token —; follow-up → "The current balance for account **acme-42** is **$1,234.00**. ## Calc"

Result

{
  "signatureType": "anthropic",
  "tokenCaptured": false,
  "reasoned": false,
  "answered": true,
  "continuation": "The current balance for account **acme-42** is **$1,234.00**.\n\n## Calculation\n\nUsing compound interest: **FV = P × (1 + r)ⁿ**\n\n- P = $1,234\n"
}

c92e8325-e9f8-4717-b95d-69e7762d32cb

Jul 6, 2026, 21:57:10

anthropic:claude-fable-5

Input

523

Output

53

Reasoning

0

Answer

53

Cost

$0.0079

Mode

streaming

Cost breakdown

Input $0.0052Output $0.0027

Latency

4.9s

Tok/s

10.8

86754ea6-23fc-40c6-bb51-fdfa14e8ff73

Jul 6, 2026, 21:57:19

anthropic:claude-fable-5

Input

593

Output

676

Reasoning

263

Answer

413

Cost

$0.0397

Mode

streaming

Cost breakdown

Input $0.0059Output $0.0338

Latency

9.9s

TTFT

9.9s

Tok/s

68.2

Tok/s (adj)

40411.0

prompt-cache

ok

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

anthropic cache; cacheRead per attempt [14458]; hit=yes.

Result

{
  "cacheStyle": "anthropic",
  "cacheReadPerAttempt": [
    14458
  ],
  "cacheReadTokens": 14458,
  "cacheHitCost": 0.018668
}

65bb10b2-176b-482d-a8d7-5194263f0cf3

Jul 6, 2026, 21:57:24

anthropic:claude-fable-5

Input

14,479

Output

80

Cache Write

14,458

Cost

$0.1849

Mode

non-streaming

Cost breakdown

Input $0.000210Output $0.0040Cache write $0.1807

Latency

4.8s

Tok/s

16.6

367be750-0613-43e7-b2d8-040d90f14820

Jul 6, 2026, 21:57:32

anthropic:claude-fable-5

Input

14,479

Output

80

Cache Read

14,458

Cost

$0.0187

Mode

non-streaming

Cost breakdown

Input $0.000210Output $0.0040Cache read $0.0145

Latency

7.4s

Tok/s

10.9

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
}

fa2ed1f7-4ce7-4273-b833-f46d536f462c

Jul 6, 2026, 21:57:35

anthropic:claude-fable-5

Input

86

Output

55

Reasoning

15

Answer

40

Cost

$0.0036

Mode

streaming

Cost breakdown

Input $0.000860Output $0.0028

Latency

3.6s

TTFT

3.1s

Tok/s

15.2

Tok/s (adj)

86.1

summarize-memex

ok

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

Read ~14533 input tokens, wrote 512 words (1167 output tokens) in 19914ms.

Result

{
  "wordCount": 512,
  "inputTokens": 14533,
  "outputTokens": 1167,
  "latencyMs": 19914
}

0fd0c81f-ae35-442c-9b42-f7498d9faf45

Jul 6, 2026, 21:57:55

anthropic:claude-fable-5

Input

14,533

Output

1,167

Reasoning

47

Answer

1,120

Cost

$0.2037

Mode

streaming

Cost breakdown

Input $0.1453Output $0.0584

Latency

19.9s

TTFT

3.8s

Tok/s

58.6

Tok/s (adj)

69.7

truncation

ok

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

finishReason=length, out=100 tokens.

Result

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

a7c79e9d-fb39-4cbd-95ad-69d7d1c14d4a

Jul 6, 2026, 21:58:02

anthropic:claude-fable-5

Input

30

Output

100

Reasoning

42

Answer

58

Cost

$0.0053

Mode

streaming

Cost breakdown

Input $0.000300Output $0.0050

Latency

6.7s

TTFT

4.4s

Tok/s

14.8

Tok/s (adj)

25.1

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"
}