VerdictPal
Join

Models / Alibaba

Qwen3.5 397B A17B (Reasoning)

Qwen3.5 397B · Released 2026-02-16

AA Intelligence Index 33.7. Listed on Artificial Analysis with API pricing $0.6/$3.6 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#140 of 184 on AA Index · current snapshot
19 AA Index · 10 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite50
ContextUnlisted
Input / 1M$0.60
Output / 1M$3.6
Knowledge cutoffunpublished
Output speed84 t/s
TTFT (default API)1.55s
Statussolid

Family profile

Best published score in each covered benchmark family.

66/100 avg
Reasoning
89
Coding
45
Agentic
51
Long context
77

4 tested benchmark families

Benchmark placements

Where Qwen3.5 397B A17B (Reasoning) places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
IFBenchReasoning#10/ 12779Artificial Analysis2026-09-10
AA output speedPerformance#44/ 190Artificial Analysis2026-09-10
AA time to first tokenPerformance#45/ 190Artificial Analysis2026-09-10
GPQA DiamondReasoning#56/ 19089Artificial Analysis2026-09-10
SciCodeCoding#62/ 9345Artificial Analysis2026-09-10
AA-LCRLong context#67/ 18277Artificial Analysis2026-09-10
τ³-Bench BankingAgentic#85/ 11013Artificial Analysis2026-09-10
Humanity's Last ExamReasoning#86/ 18529Artificial Analysis2026-09-10
Terminal-BenchAgentic#93/ 18951Artificial Analysis2026-09-10
Artificial Analysis Intelligence IndexReasoning#144/ 18819Artificial Analysis2026-09-10
Family context

The three highest-scoring models with pages in each capability family. Where Qwen3.5 397B A17B (Reasoning) shows up, it's highlighted.

ReasoningQwen3.5 397B A17B (Reasoning) · 89
  1. 1
    GPT-6 Astra
    96
  2. 2
    Claude Mythos Preview
    95
  3. 3
    Gemini 3.7 Flash
    95
CodingQwen3.5 397B A17B (Reasoning) · 45
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticQwen3.5 397B A17B (Reasoning) · 51
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextQwen3.5 397B A17B (Reasoning) · 77
  1. 1
    Kimi K3
    89
  2. 2
    Claude Fable 5.1
    85
  3. 3
    GPT-5.5 (high)
    84
Newest receipts

Benchmark rows added to the public ledger for Qwen3.5 397B A17B (Reasoning) in the last 120 days. Older rows live in the full table below.

10 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context77.33%Artificial Analysis6
AA output speedPerformance84 t/sArtificial Analysis6
AA time to first tokenPerformance1.55sArtificial Analysis6
Artificial Analysis Intelligence IndexReasoning19.1%Artificial Analysis6
GPQA DiamondReasoning89.3%Artificial Analysis6
Humanity's Last ExamReasoning29%Artificial Analysis6
IFBenchReasoning78.78%Artificial Analysis6
SciCodeCoding44.8%Artificial Analysis6
τ³-Bench BankingAgentic13.4%Artificial Analysis6
Terminal-BenchAgentic51.31%Artificial Analysis6
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning89
Coding45
Agentic51
Long context77
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaxAIMiniMaxZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Qwen3.5 397B A17B (Reasoning). Scores link to the original source; gaps mean no public row exists yet.

10 sourced rows

10 of 37 catalog benchmarks have a sourced row for Qwen3.5 397B A17B (Reasoning).

Reasoning

Coding

Agentic

Long context

Performance

Sources

The Pack · Editorial newsletter

New tools in your inbox. Free.

One short email when a tool ships or changes status. No tracking, no third-party analytics. Unsubscribe in one click.