Armand McQueen

basic-generation

ok

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

Produced 99 words (149 output tokens).

Result

{
  "wordCount": 99,
  "outputTokens": 149
}

26d944da-3173-4b88-a623-9d4f3b6816bb

Aug 8, 2026, 08:53:33

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

Input

15

Output

149

Reasoning

0

Answer

149

Cost

$0.000044

Mode

streaming

Cost breakdown

Input $0.000002Output $0.000042

Latency

2.3s

TTFT

527ms

Tok/s

65.5

Tok/s (adj)

85.2

usage-cost

ok

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

in=14 out=192 rt=185 cost=$0.000056 (paid).

Result

{
  "inputTokens": 14,
  "outputTokens": 192,
  "reasoningTokens": 185,
  "costUsd": 0.00005572000000000001,
  "paid": true
}

59524adb-998b-499a-b70f-3915e3f5f189

Aug 8, 2026, 08:53:36

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

Input

14

Output

192

Reasoning

185

Answer

7

Cost

$0.000056

Mode

streaming

Cost breakdown

Input $0.000002Output $0.000054

Latency

2.7s

TTFT

2.7s

Tok/s

70.6

Tok/s (adj)

287.4

streaming-structure

ok

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

mode=streaming, 34 stream events, 1 text block(s), 35 chars.

Result

{
  "mode": "streaming",
  "streamEventCount": 34,
  "textBlocks": 1,
  "textLen": 35
}

5391fc9c-9be9-40f3-abb8-66d2124d05b8

Aug 8, 2026, 08:53:38

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

Input

17

Output

111

Reasoning

80

Answer

31

Cost

$0.000033

Mode

streaming

Cost breakdown

Input $0.000002Output $0.000031

Latency

2.2s

TTFT

2.1s

Tok/s

51.1

Tok/s (adj)

293.8

complete-stream-parity

ok

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

stream="Paris" · complete="Paris"

Result

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

ed2260c0-2a83-49f2-9bf3-adef05838077

Aug 8, 2026, 08:53:38

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

Input

18

Output

2

Reasoning

0

Answer

2

Cost

$0.000003

Mode

streaming

Cost breakdown

Input $0.000003Output $0.000001

Latency

617ms

TTFT

594ms

Tok/s

3.2

Tok/s (adj)

88.4

acde40d3-e7f5-44d2-8adf-89ca64688ba7

Aug 8, 2026, 08:53:39

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

Input

18

Output

2

Reasoning

0

Answer

2

Cost

$0.000003

Mode

non-streaming

Cost breakdown

Input $0.000003Output $0.000001

Latency

492ms

Tok/s

4.1

reasoning-conformance

ok

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

Class (from reasoning-config): toggleable.
off: out=422 · mid: out=592 (rt=272).

Result

{
  "class": "toggleable",
  "mode": "deterministic",
  "offOutputTokens": 422,
  "midOutputTokens": 592,
  "midReasoningTokens": 272
}

b536521f-f893-4ad1-b378-7f813dfcefbe

Aug 8, 2026, 08:53:43

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

Input

89

Output

422

Reasoning

0

Answer

422

Cost

$0.000131

Mode

streaming

Cost breakdown

Input $0.000012Output $0.000118

Latency

3.7s

TTFT

848ms

Tok/s

115.2

Tok/s (adj)

149.9

78674447-9dc2-4ca6-aad4-e7bb6ede5153

Aug 8, 2026, 08:53:46

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

Input

89

Output

592

Reasoning

272

Answer

320

Cost

$0.000178

Mode

streaming

Cost breakdown

Input $0.000012Output $0.000166

Latency

3.7s

TTFT

1.9s

Tok/s

158.3

Tok/s (adj)

174.9

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
}

1adc9f12-c4e7-4abb-a945-99053e366fe3

Aug 8, 2026, 08:53:47

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

Input

355

Output

71

Reasoning

0

Answer

71

Cost

$0.000070

Mode

streaming

Cost breakdown

Input $0.000050Output $0.000020

Latency

950ms

TTFT

730ms

Tok/s

74.7

Tok/s (adj)

323.3

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
}

8f4fa332-6684-47b8-a4a7-086d2fa456b4

Aug 8, 2026, 08:53:48

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

Input

351

Output

19

Reasoning

0

Answer

19

Cost

$0.000054

Mode

streaming

Cost breakdown

Input $0.000049Output $0.000005

Latency

540ms

Tok/s

35.2

dc6d6f7e-2fdb-4372-bfd1-c4dc43992925

Aug 8, 2026, 08:53:49

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

Input

427

Output

15

Reasoning

0

Answer

15

Cost

$0.000064

Mode

streaming

Cost breakdown

Input $0.000060Output $0.000004

Latency

688ms

TTFT

687ms

Tok/s

21.8

Tok/s (adj)

15789.5

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 → "The current balance for account **acme-42** is **$1,234**. Now, let's"

Result

{
  "signatureType": "none",
  "tokenCaptured": false,
  "reasoned": true,
  "answered": true,
  "continuation": "The current balance for account **acme-42** is **$1,234**.\n\nNow, let's compute the balance after 7 years with 5% annual compound interest:\n\n"
}

c95585d5-4823-42a6-a65f-d3f4a9c1279c

Aug 8, 2026, 08:53:49

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

Input

381

Output

108

Reasoning

58

Answer

50

Cost

$0.000084

Mode

streaming

Cost breakdown

Input $0.000053Output $0.000030

Latency

724ms

Tok/s

149.2

fc3aa3d0-faf9-46cf-9c89-13c971587fdc

Aug 8, 2026, 08:53:51

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

Input

448

Output

325

Reasoning

0

Answer

325

Cost

$0.000154

Mode

streaming

Cost breakdown

Input $0.000063Output $0.000091

Latency

2.2s

TTFT

301ms

Tok/s

147.9

Tok/s (adj)

171.3

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
}

22347015-3469-4c21-9cd5-f7e6a7964dbf

Aug 8, 2026, 08:53:52

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

Input

63

Output

24

Reasoning

0

Answer

24

Cost

$0.000016

Mode

streaming

Cost breakdown

Input $0.000009Output $0.000007

Latency

673ms

TTFT

595ms

Tok/s

35.7

Tok/s (adj)

306.2

summarize-memex

ok

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

Read ~10232 input tokens, wrote 512 words (636 output tokens) in 9452ms.

Result

{
  "wordCount": 512,
  "inputTokens": 10232,
  "outputTokens": 636,
  "latencyMs": 9452
}

e70d0edf-94a7-475c-a9e7-ea8ffae55a77

Aug 8, 2026, 08:54:02

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

Input

10,232

Output

636

Reasoning

0

Answer

636

Cost

$0.0016

Mode

streaming

Cost breakdown

Input $0.0014Output $0.000178

Latency

9.5s

TTFT

1.1s

Tok/s

67.3

Tok/s (adj)

76.2

truncation

ok

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

finishReason=length, out=100 tokens.

Result

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

cd602648-02bd-4caa-9b41-c7408d7ab85d

Aug 8, 2026, 08:54:04

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

Input

21

Output

100

Reasoning

0

Answer

100

Cost

$0.000031

Mode

streaming

Cost breakdown

Input $0.000003Output $0.000028

Latency

2.4s

TTFT

737ms

Tok/s

41.7

Tok/s (adj)

60.3

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"
}