VerdictPal · editorial desk · updated 1 Sep 2026VerdictPal

Compare · Two models, head-to-head

Pick two models. See the diff.

Composite scores, benchmark ranks, capability grids, and pricing per million tokens. All sourced from the same data that powers the model atlas.

Tools Models

Qwen3.8-Flash-Next vs Qwen3.8 27B

Model A

Model B

MODEL A

Qwen3.8-Flash-Next

Alibaba · Qwen

Composite 63Intel 569 rows1 sources

Alibaba's 26 Aug 2026 Flash-Next sibling in the Qwen3.8 line. Artificial Analysis Intelligence Index 55.8. AA hosted list $0.15 / $0.47 per 1M tokens — cheaper and faster than Qwen3.8 27B, below hosted Qwen3.8 Max.

MODEL B

Qwen3.8 27B

Alibaba · Qwen

Composite 60Intel 529 rows1 sources

Alibaba's dense open-weight Qwen3.8 sibling (14 Aug 2026). Artificial Analysis Intelligence Index 52 on the xhigh-effort endpoint. AA hosted list $0.50 / $3 per 1M tokens. Apache 2.0 weights on Hugging Face.

Qwen3.8-Flash-Next4stats won

Tied2stats

Qwen3.8 27B1stats won

Spec showdown

Capacity, price, freshness.

Qwen3.8-Flash-NextvsQwen3.8 27B

Composite scores are within 5 points — 63 for Qwen3.8-Flash-Next vs 60 for Qwen3.8 27B. The choice comes down to capability profile and price. Qwen3.8-Flash-Next has the higher Vals/AA intelligence score (56 vs 52). Qwen3.8-Flash-Next is significantly cheaper at $0.15/1M input vs $0.50/1M input.

Composite score
Qwen3.8-Flash-Next63Winner on this rowQwen3.8 27B60
Intelligence (Vals/AA)
Qwen3.8-Flash-Next56Winner on this rowQwen3.8 27B52
Context window
Qwen3.8-Flash-Next262KQwen3.8 27B262K
Max output
Qwen3.8-Flash-Next—Qwen3.8 27B—
Input price / 1M (lower wins)
Qwen3.8-Flash-Next$0.15Winner on this rowQwen3.8 27B$0.50
Output speed (AA default API)
Qwen3.8-Flash-Next87 t/sWinner on this rowQwen3.8 27B45 t/s
Time to first token (AA default API)
Qwen3.8-Flash-Next1.27sQwen3.8 27B1.21sWinner on this row

Composite score

Intelligence (Vals/AA)

Context window

262K

Max output

—

Input price / 1M (lower wins)

$0.15

$0.50

Output speed (AA default API)

87 t/s

45 t/s

Time to first token (AA default API)

1.27s

1.21s

Capability grid

What each model can do.

Modalities

Qwen3.8-Flash-NextQwen3.8 27B

Text●●

Image—●

Audio——

Video—●

Code●●

Tool use●●

Reasoning●●

Access

Qwen3.8-Flash-NextQwen3.8 27B

API●●

Consumer app——

Open weights●●

Managed cloud——

On-device——

Family profile

Top score per benchmark family.

Reasoning

Coding

Agentic

Long context

Benchmark ranks

Every shared benchmark, side by side.

Reasoning

Artificial Analysis Intelligence Index#20/184 · #31/184

55.8%

52%

GPQA Diamond#21/185 · #39/185

92.3%

90.5%

Humanity's Last Exam#42/180 · #57/180

38%

33.9%

Coding

SciCode#67/180 · #84/180

46.9%

44.7%

Agentic

Terminal-Bench#8/184 · #26/184

86.14%

79.78%

τ³-Bench Banking#8/106 · #5/106

45.36%

48.04%

Long context

AA-LCR#30/179 · #25/179

77%

77.33%

Performance

AA output speed#46/185 · #73/185

87 t/s

45 t/s

AA time to first token#54/185 · #57/185

1.27s

1.21s

Dossier facts

Dates, sources, fineprint.

Qwen3.8-Flash-NextQwen3.8 27B

ProviderAlibabaAlibaba

FamilyQwenQwen

Release date2026-08-262026-08-14

Knowledge cutoffunpublishedunpublished

Context window262K262K

Max output——

Input price /1M$0.15$0.50

Output price /1M$0.47$3.0

Statussolidsolid

Composite score6360

Intelligence score5652

Intel sourceaaaa

Benchmark rows99

Source count11

Freshnessfreshfresh

Pricing checked2026-09-012026-08-24

Ecosystem

Which tools wrap each model.

Qwen3.8-Flash-Next

Not listed in any tool yet.

Qwen3.8 27B

Not listed in any tool yet.

Verdict

Which model to actually pick.

Open Qwen3.8-Flash-Next's dossier Open Qwen3.8 27B's dossier