gpt-5.6-sol

7/11/2026, 8:15:44 PM · 15 tests · run 59327c98-4c4a-442b-beac-6ce5236c83da

basic-generation

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

Produced 100 words (140 output tokens).

Result

{
  "wordCount": 100,
  "outputTokens": 140
}

36abde4f-2bcb-46d6-bfef-fbea6882513f

Jul 11, 2026, 20:14:57

openai:gpt-5.6-sol

Input

Output

140

Reasoning

Answer

140

Cost

$0.0043

Mode

streaming

Cost breakdown

Input $0.000085Output $0.0042

Latency

3.0s

TTFT

713ms

Tok/s

47.3

Tok/s (adj)

62.4

usage-cost

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

in=14 out=9 rt=0 cost=$0.000340 (paid).

Result

{
  "inputTokens": 14,
  "outputTokens": 9,
  "reasoningTokens": 0,
  "costUsd": 0.00034,
  "paid": true
}

b5aad8a6-f2a4-43e1-8485-48d86a749fdb

Jul 11, 2026, 20:14:58

openai:gpt-5.6-sol

Input

Output

Reasoning

Answer

Cost

$0.000340

Mode

streaming

Cost breakdown

Input $0.000070Output $0.000270

Latency

924ms

TTFT

766ms

Tok/s

9.7

Tok/s (adj)

57.1

streaming-structure

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

mode=streaming, 37 stream events, 1 text block(s), 35 chars.

Result

{
  "mode": "streaming",
  "streamEventCount": 37,
  "textBlocks": 1,
  "textLen": 35
}

7fa65406-a946-47a1-89d0-4fed3ce553c3

Jul 11, 2026, 20:14:59

openai:gpt-5.6-sol

Input

Output

Reasoning

Answer

Cost

$0.0014

Mode

streaming

Cost breakdown

Input $0.000095Output $0.0013

Latency

1.2s

TTFT

751ms

Tok/s

37.8

Tok/s (adj)

85.0

complete-stream-parity

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

stream="Paris" · complete="Paris"

Result

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

56689f8e-2508-4b92-859b-aa28f0eebae0

Jul 11, 2026, 20:15:00

openai:gpt-5.6-sol

Input

Output

Reasoning

Answer

Cost

$0.000250

Mode

streaming

Cost breakdown

Input $0.000100Output $0.000150

Latency

890ms

TTFT

668ms

Tok/s

5.6

Tok/s (adj)

22.5

d717ead4-1587-42f2-b153-2efba1a567fc

Jul 11, 2026, 20:15:03

openai:gpt-5.6-sol

Input

Output

Reasoning

Answer

Cost

$0.000250

Mode

non-streaming

Cost breakdown

Input $0.000100Output $0.000150

Latency

2.8s

Tok/s

1.8

reasoning-conformance

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

Class (from reasoning-config): toggleable.
off: out=246 · mid: out=319 (rt=81).

Result

{
  "class": "toggleable",
  "mode": "deterministic",
  "offOutputTokens": 246,
  "midOutputTokens": 319,
  "midReasoningTokens": 81
}

c8329a96-5061-4626-84c0-f9eb32696ced

Jul 11, 2026, 20:15:06

openai:gpt-5.6-sol

Input

Output

246

Reasoning

Answer

246

Cost

$0.0078

Mode

streaming

Cost breakdown

Input $0.000455Output $0.0074

Latency

3.2s

TTFT

591ms

Tok/s

78.0

Tok/s (adj)

96.0

d29adf7d-c2f0-421f-a4d4-be34aca16e60

Jul 11, 2026, 20:15:10

openai:gpt-5.6-sol

Input

Output

319

Reasoning

Answer

238

Cost

$0.0100

Mode

streaming

Cost breakdown

Input $0.000455Output $0.0096

Latency

3.9s

TTFT

3.4s

Tok/s

81.4

Tok/s (adj)

469.9

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
}

f121e1f4-8ac6-4c93-98d0-a72b2919942d

Jul 11, 2026, 20:15:11

openai:gpt-5.6-sol

Input

Output

Reasoning

Answer

Cost

$0.0011

Mode

streaming

Cost breakdown

Input $0.000425Output $0.000630

Latency

880ms

Tok/s

23.9

tool-loop

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 and 17 is 391."

Result

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

e4d5955f-c4c1-43ad-b479-a2e40d7a9ea2

Jul 11, 2026, 20:15:12

openai:gpt-5.6-sol

Input

Output

Reasoning

Answer

Cost

$0.0011

Mode

streaming

Cost breakdown

Input $0.000440Output $0.000630

Latency

920ms

Tok/s

22.8

fc962831-0031-4018-8aac-162d30de5964

Jul 11, 2026, 20:15:13

openai:gpt-5.6-sol

Input

124

Output

Reasoning

Answer

Cost

$0.0011

Mode

streaming

Cost breakdown

Input $0.000620Output $0.000480

Latency

886ms

TTFT

614ms

Tok/s

18.1

Tok/s (adj)

58.9

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.

Step 1 didn't engage reasoning (adaptive discretion) — signature not exercised this run.
reasoned=false, openai token —; follow-up → "The current balance is **$1,234**. Calculation after 7 years at 5% an"

Result

{
  "signatureType": "openai",
  "tokenCaptured": false,
  "reasoned": false,
  "answered": true,
  "continuation": "The current balance is **$1,234**.\n\nCalculation after 7 years at 5% annual compound growth:\n\n\\[\n\\$1{,}234 \\times (1.05)^7\n= \\$1{,}234 \\times"
}

45d8d813-63e5-4c63-a462-93965460f8e0

Jul 11, 2026, 20:15:16

openai:gpt-5.6-sol

Input

124

Output

Reasoning

Answer

Cost

$0.0013

Mode

streaming

Cost breakdown

Input $0.000620Output $0.000660

Latency

911ms

Tok/s

24.1

12220fd4-81f2-4e13-9c99-be40570eb2f8

Jul 11, 2026, 20:15:18

openai:gpt-5.6-sol

Input

164

Output

134

Reasoning

Answer

134

Cost

$0.0048

Mode

streaming

Cost breakdown

Input $0.000820Output $0.0040

Latency

2.0s

TTFT

534ms

Tok/s

67.1

Tok/s (adj)

91.6

prompt-cache

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

implicit cache; cacheRead per attempt [10173]; hit=yes.

Result

{
  "cacheStyle": "implicit",
  "cacheReadPerAttempt": [
    10173
  ],
  "cacheReadTokens": 10173,
  "cacheHitCost": 0.0061515
}

47f7cf78-1335-4ab9-979c-24dc88f85ede

Jul 11, 2026, 20:15:20

openai:gpt-5.6-sol

Input

10,176

Output

Reasoning

Answer

Cost

$0.0519

Mode

non-streaming

Cost breakdown

Input $0.0509Output $0.0010

Latency

1.8s

Tok/s

19.9

fd03758b-6a0d-4932-b685-f55697358735

Jul 11, 2026, 20:15:22

openai:gpt-5.6-sol

Input

10,176

Output

Reasoning

Answer

Cache Read

10,173

Cost

$0.0062

Mode

non-streaming

Cost breakdown

Input $0.000015Output $0.0010Cache read $0.0051

Latency

1.5s

Tok/s

22.8

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

fc419071-464e-47dd-ba2a-096c9a82b6a7

Jul 11, 2026, 20:15:22

openai:gpt-5.6-sol

Input

Output

Reasoning

Answer

Cost

$0.0011

Mode

streaming

Cost breakdown

Input $0.000335Output $0.000810

Latency

955ms

TTFT

567ms

Tok/s

28.3

Tok/s (adj)

69.6

summarize-memex

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

Read ~10208 input tokens, wrote 553 words (704 output tokens) in 17490ms.

Result

{
  "wordCount": 553,
  "inputTokens": 10208,
  "outputTokens": 704,
  "latencyMs": 17490
}

6131164c-36ce-4a08-ad94-f2dfeab333f4

Jul 11, 2026, 20:15:40

openai:gpt-5.6-sol

Input

10,208

Output

704

Reasoning

Answer

704

Cost

$0.0722

Mode

streaming

Cost breakdown

Input $0.0510Output $0.0211

Latency

17.5s

TTFT

1.5s

Tok/s

40.3

Tok/s (adj)

44.1

truncation

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

finishReason=length, out=100 tokens.

Result

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

c38847b1-c944-4e9e-bb93-07868ff47a83

Jul 11, 2026, 20:15:43

openai:gpt-5.6-sol

Input

Output

100

Reasoning

Answer

100

Cost

$0.0031

Mode

streaming

Cost breakdown

Input $0.000115Output $0.0030

Latency

2.7s

TTFT

760ms

Tok/s

36.4

Tok/s (adj)

50.3

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"
}