Armand McQueen

basic-generation

ok

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

Produced 88 words (110 output tokens).

Result

{
  "wordCount": 88,
  "outputTokens": 110
}

2f2512ab-7da7-47de-ae7a-7f0b96d5c9b0

Sep 5, 2026, 02:35:26

togetherai:Qwen/Qwen3.8-Flash

Input

38

Output

110

Reasoning

0

Answer

110

Cost

$0.000057

Mode

streaming

Cost breakdown

Input $0.000006Output $0.000052

Latency

3.5s

TTFT

1.5s

Tok/s

31.0

Tok/s (adj)

54.2

usage-cost

ok

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

in=69 out=128 rt=120 cost=$0.000071 (paid).

Result

{
  "inputTokens": 69,
  "outputTokens": 128,
  "reasoningTokens": 120,
  "costUsd": 0.00007050999999999999,
  "paid": true
}

18efcc2f-d936-438c-b5e7-b62b4b31b487

Sep 5, 2026, 02:35:30

togetherai:Qwen/Qwen3.8-Flash

Input

69

Output

128

Reasoning

120

Answer

8

Cost

$0.000071

Mode

streaming

Cost breakdown

Input $0.000010Output $0.000060

Latency

3.9s

TTFT

3.9s

Tok/s

32.5

Tok/s (adj)

93.3

streaming-structure

ok

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

mode=streaming, 28 stream events, 1 text block(s), 35 chars.

Result

{
  "mode": "streaming",
  "streamEventCount": 28,
  "textBlocks": 1,
  "textLen": 35
}

667c82de-411a-4eda-869a-50b4f20696c9

Sep 5, 2026, 02:35:32

togetherai:Qwen/Qwen3.8-Flash

Input

75

Output

72

Reasoning

34

Answer

38

Cost

$0.000045

Mode

streaming

Cost breakdown

Input $0.000011Output $0.000034

Latency

1.9s

TTFT

1.6s

Tok/s

38.7

Tok/s (adj)

170.5

complete-stream-parity

ok

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

Test threw: This model only supports streaming. Set "stream": true.

Result

null

reasoning-conformance

ok

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

Class (from reasoning-config): toggleable.
off: out=875 · mid: out=1157 (rt=615, rchars=1233).

Result

{
  "class": "toggleable",
  "mode": "deterministic",
  "offOutputTokens": 875,
  "midOutputTokens": 1157,
  "midReasoningTokens": 615
}

24c7bda0-e49e-4013-96fd-d870ad1bfba3

Sep 5, 2026, 02:35:44

togetherai:Qwen/Qwen3.8-Flash

Input

116

Output

875

Reasoning

0

Answer

875

Cost

$0.000429

Mode

streaming

Cost breakdown

Input $0.000017Output $0.000411

Latency

11.6s

TTFT

1.9s

Tok/s

75.2

Tok/s (adj)

89.8

906909dc-85aa-4bcb-8070-fcb4e28febeb

Sep 5, 2026, 02:35:55

togetherai:Qwen/Qwen3.8-Flash

Input

114

Output

1,157

Reasoning

615

Answer

542

Cost

$0.000561

Mode

streaming

Cost breakdown

Input $0.000017Output $0.000544

Latency

10.4s

TTFT

6.3s

Tok/s

111.6

Tok/s (adj)

133.8

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
}

63873d16-54ec-4b41-bd89-776e090cede9

Sep 5, 2026, 02:35:57

togetherai:Qwen/Qwen3.8-Flash

Input

348

Output

37

Reasoning

0

Answer

37

Cost

$0.000070

Mode

streaming

Cost breakdown

Input $0.000052Output $0.000017

Latency

2.2s

Tok/s

17.1

tool-loop

ok

Forces a tool call, feeds the result back, and checks the continuation uses it.

step1 (auto): 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": false
}

44da3c4b-8f53-4cf1-8c9d-d2506bb04595

Sep 5, 2026, 02:35:59

togetherai:Qwen/Qwen3.8-Flash

Input

380

Output

65

Reasoning

25

Answer

40

Cost

$0.000088

Mode

streaming

Cost breakdown

Input $0.000057Output $0.000031

Latency

1.7s

Tok/s

37.7

b2c73ea2-ce3f-4581-8d91-3b0eba9b4f87

Sep 5, 2026, 02:36:01

togetherai:Qwen/Qwen3.8-Flash

Input

468

Output

37

Reasoning

18

Answer

19

Cost

$0.000088

Mode

streaming

Cost breakdown

Input $0.000070Output $0.000017

Latency

1.2s

TTFT

1.1s

Tok/s

30.9

Tok/s (adj)

138.7

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 lookup:** Account `acme-42` currently holds **$1,234.00**. "

Result

{
  "signatureType": "none",
  "tokenCaptured": false,
  "reasoned": true,
  "answered": true,
  "continuation": "**Balance lookup:** Account `acme-42` currently holds **$1,234.00**.\n\n**Growth calculation (5% compounded annually for 7 years):**\n\nFuture v"
}

3d261032-df19-4817-a271-5673f3dfad60

Sep 5, 2026, 02:36:02

togetherai:Qwen/Qwen3.8-Flash

Input

368

Output

41

Reasoning

9

Answer

32

Cost

$0.000074

Mode

streaming

Cost breakdown

Input $0.000055Output $0.000019

Latency

1.7s

Tok/s

24.1

2a9eaffe-a3f5-43af-b539-aad6225eb4d3

Sep 5, 2026, 02:36:09

togetherai:Qwen/Qwen3.8-Flash

Input

437

Output

340

Reasoning

0

Answer

340

Cost

$0.000225

Mode

streaming

Cost breakdown

Input $0.000066Output $0.000160

Latency

6.6s

TTFT

1.9s

Tok/s

51.6

Tok/s (adj)

73.2

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

46e67c3e-ccee-413d-a72e-eeef6bcce919

Sep 5, 2026, 02:36:11

togetherai:Qwen/Qwen3.8-Flash

Input

74

Output

27

Reasoning

0

Answer

27

Cost

$0.000024

Mode

streaming

Cost breakdown

Input $0.000011Output $0.000013

Latency

2.0s

TTFT

1.7s

Tok/s

13.8

Tok/s (adj)

90.9

summarize-memex

ok

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

Read ~10388 input tokens, wrote 541 words (681 output tokens) in 9494ms.

Result

{
  "wordCount": 541,
  "inputTokens": 10388,
  "outputTokens": 681,
  "latencyMs": 9494
}

406046f8-a0c8-4a87-bed7-19f67b4772c1

Sep 5, 2026, 02:36:20

togetherai:Qwen/Qwen3.8-Flash

Input

10,388

Output

681

Reasoning

0

Answer

681

Cost

$0.0019

Mode

streaming

Cost breakdown

Input $0.0016Output $0.000320

Latency

9.5s

TTFT

2.8s

Tok/s

71.7

Tok/s (adj)

101.8

truncation

ok

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

finishReason=length, out=100 tokens.

Result

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

c91e8c45-4812-4b75-88f3-2768bd7c9487

Sep 5, 2026, 02:36:22

togetherai:Qwen/Qwen3.8-Flash

Input

44

Output

100

Reasoning

0

Answer

100

Cost

$0.000054

Mode

streaming

Cost breakdown

Input $0.000007Output $0.000047

Latency

2.1s

TTFT

914ms

Tok/s

48.5

Tok/s (adj)

87.1

streaming-only-rejection

ok

Streaming-only models reject complete(); all others accept it.

streamingOnly=false; complete() threw (This model only supports streaming. Set "stream": true.).

Result

{
  "streamingOnly": false,
  "completeThrew": true
}

abort

ok

A pre-aborted signal rejects the call (abort is threaded + honored).

pre-aborted signal → rejected (AbortError).

Result

{
  "rejected": true,
  "errorName": "AbortError"
}