Armand McQueen

basic-generation

ok

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

Produced 119 words (169 output tokens).

Result

{
  "wordCount": 119,
  "outputTokens": 169
}

3c8eaa82-2e79-4fc8-b9a0-a73a806e14b7

Aug 18, 2026, 08:15:10

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

15

Output

169

Reasoning

0

Answer

169

Cost

$0.000032

Mode

streaming

Cost breakdown

Input $0.000001Output $0.000030

Latency

22.8s

TTFT

5.2s

Tok/s

7.4

Tok/s (adj)

9.6

usage-cost

ok

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

in=14 out=6 rt=0 cost=$0.000002 (paid).

Result

{
  "inputTokens": 14,
  "outputTokens": 6,
  "reasoningTokens": 0,
  "costUsd": 0.0000022,
  "paid": true
}

dec5dd98-6e46-488b-99b1-a67db365fb1d

Aug 18, 2026, 08:15:15

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

14

Output

6

Reasoning

0

Answer

6

Cost

$0.000002

Mode

streaming

Cost breakdown

Input $0.000001Output $0.000001

Latency

5.3s

TTFT

5.3s

Tok/s

1.1

Tok/s (adj)

5128.2

streaming-structure

ok

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

mode=streaming, 10 stream events, 1 text block(s), 63 chars.

Result

{
  "mode": "streaming",
  "streamEventCount": 10,
  "textBlocks": 1,
  "textLen": 63
}

00d62646-3be4-4403-b4e1-0d0c538b6cb1

Aug 18, 2026, 08:15:16

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

17

Output

30

Reasoning

0

Answer

30

Cost

$0.000007

Mode

streaming

Cost breakdown

Input $0.000001Output $0.000005

Latency

1.2s

TTFT

971ms

Tok/s

25.3

Tok/s (adj)

140.4

complete-stream-parity

ok

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

stream="Paris" · complete="Paris"

Result

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

b2c13113-b13b-47f3-8e73-7e2321a42a80

Aug 18, 2026, 08:15:17

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

18

Output

2

Reasoning

0

Answer

2

Cost

$0.000002

Mode

streaming

Cost breakdown

Input $0.000001Output $0.000000

Latency

362ms

TTFT

360ms

Tok/s

5.5

Tok/s (adj)

1036.3

00d34be0-fc1a-4e9d-b9b6-03cf63f63489

Aug 18, 2026, 08:15:17

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

18

Output

2

Reasoning

0

Answer

2

Cost

$0.000002

Mode

non-streaming

Cost breakdown

Input $0.000001Output $0.000000

Latency

420ms

Tok/s

4.8

reasoning-conformance

ok

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

Class (from reasoning-config): toggleable.
off: out=425 · mid: out=672 (rt=0, rchars=942).

Result

{
  "class": "toggleable",
  "mode": "deterministic",
  "offOutputTokens": 425,
  "midOutputTokens": 672,
  "midReasoningTokens": 0
}

d0273728-fb4d-4640-8d74-de89af72d2e2

Aug 18, 2026, 08:15:42

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

89

Output

425

Reasoning

0

Answer

425

Cost

$0.000084

Mode

streaming

Cost breakdown

Input $0.000007Output $0.000076

Latency

25.1s

TTFT

271ms

Tok/s

16.9

Tok/s (adj)

17.1

45e17b70-6a14-4533-b49d-906234b4daf3

Aug 18, 2026, 08:16:09

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

89

Output

672

Reasoning

0

Answer

672

Cost

$0.000128

Mode

streaming

Cost breakdown

Input $0.000007Output $0.000121

Latency

26.2s

TTFT

17.5s

Tok/s

25.6

Tok/s (adj)

77.0

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
}

871d904b-062f-436d-88f1-ee6fe32cd19c

Aug 18, 2026, 08:16:10

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

351

Output

60

Reasoning

0

Answer

60

Cost

$0.000039

Mode

streaming

Cost breakdown

Input $0.000028Output $0.000011

Latency

1.6s

Tok/s

37.5

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 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
}

6c5a271d-c1fa-432b-aea0-3af7d4d99fb1

Aug 18, 2026, 08:16:11

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

347

Output

60

Reasoning

0

Answer

60

Cost

$0.000039

Mode

streaming

Cost breakdown

Input $0.000028Output $0.000011

Latency

813ms

Tok/s

73.8

c55a63c9-aca4-4e8a-9899-a2048567ab9d

Aug 18, 2026, 08:16:11

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

422

Output

13

Reasoning

0

Answer

13

Cache Read

256

Cost

$0.000020

Mode

streaming

Cost breakdown

Input $0.000013Output $0.000002Cache read $0.000004

Latency

390ms

TTFT

323ms

Tok/s

33.3

Tok/s (adj)

193.3

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 → "The original balance is **$1,234**. ### Future value after 7 years at"

Result

{
  "signatureType": "none",
  "tokenCaptured": false,
  "reasoned": false,
  "answered": true,
  "continuation": "The original balance is **$1,234**.\n\n### Future value after 7 years at 5% compounded annually\n\nThe formula for compound interest is:\n\n\\[\nA ="
}

d90f35b2-a058-40aa-ac37-ed9e04e314b2

Aug 18, 2026, 08:16:17

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

377

Output

83

Reasoning

0

Answer

83

Cache Read

256

Cost

$0.000029

Mode

streaming

Cost breakdown

Input $0.000010Output $0.000015Cache read $0.000004

Latency

1.3s

Tok/s

64.6

2ce00aaf-00b6-462b-8ef9-1f81cbc645d0

Aug 18, 2026, 08:16:24

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

444

Output

445

Reasoning

0

Answer

445

Cache Read

256

Cost

$0.000099

Mode

streaming

Cost breakdown

Input $0.000015Output $0.000080Cache read $0.000004

Latency

6.8s

TTFT

764ms

Tok/s

65.9

Tok/s (adj)

74.3

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

94fabc33-c03b-48b5-8f9d-f593c81d99e4

Aug 18, 2026, 08:16:26

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

63

Output

29

Reasoning

0

Answer

29

Cost

$0.000010

Mode

streaming

Cost breakdown

Input $0.000005Output $0.000005

Latency

2.5s

TTFT

2.0s

Tok/s

11.8

Tok/s (adj)

57.0

summarize-memex

ok

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

Read ~10232 input tokens, wrote 549 words (692 output tokens) in 29663ms.

Result

{
  "wordCount": 549,
  "inputTokens": 10232,
  "outputTokens": 692,
  "latencyMs": 29663
}

961d6c84-6467-4771-87d9-2ac2f15f4484

Aug 18, 2026, 08:16:56

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

10,232

Output

692

Reasoning

0

Answer

692

Cost

$0.000943

Mode

streaming

Cost breakdown

Input $0.000819Output $0.000125

Latency

29.7s

TTFT

2.9s

Tok/s

23.3

Tok/s (adj)

25.9

truncation

ok

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

finishReason=length, out=100 tokens.

Result

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

ed2a6339-8928-47b3-88d7-33e079650bbd

Aug 18, 2026, 08:16:58

deepinfra:deepseek-ai/DeepSeek-V4-Flash-0731

Input

21

Output

100

Reasoning

0

Answer

100

Cost

$0.000020

Mode

streaming

Cost breakdown

Input $0.000002Output $0.000018

Latency

2.3s

TTFT

419ms

Tok/s

43.3

Tok/s (adj)

53.0

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"
}