VerdictPal
Join

Models / Alibaba

Qwen3.8 2.4T A95B

Qwen · Released 2026-08-12

Alibaba's open-weight Qwen-Max-class MoE (12 Aug 2026): 2.4T total / 95B active. Artificial Analysis Intelligence Index 57.7. AA list price $2 / $6 per 1M tokens.

Self-hostable Qwen3.8 checkpoint for long-context and agentic work. Hosted Qwen3.8 Max adds vision and a non-thinking mode. Prefer sourced ledger rows over vendor eval tables.

#7 of 166 on AA Index · current snapshot
58 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite64
Context262K
Input / 1M$2.0
Output / 1M$6.0
Knowledge cutoffunpublished
Output speed40 t/s
TTFT (default API)1.76s
Statussolid

Family profile

Best published score in each covered benchmark family.

76/100 avg
Reasoning
94
Coding
52
Agentic
82
Long context
75

4 tested benchmark families

Editor's note
Recorded from the 2026-08-15 AA snapshot (slug qwen3-8-2-4t-a95b, index 57.7) plus the Hugging Face model card. This is the open-weight Max-class checkpoint, not the hosted Qwen3.8 Max API product. Editorial depth still needs a desk pass.
Benchmark placements

Where Qwen3.8 2.4T A95B 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
τ³-Bench BankingAgentic#4/ 10649Artificial Analysis2026-09-01
GPQA DiamondReasoning#9/ 18594Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#12/ 18458Artificial Analysis2026-09-01
Terminal-BenchAgentic#16/ 18482Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#24/ 18042Artificial Analysis2026-09-01
AA time to first tokenPerformance#33/ 185Artificial Analysis2026-09-01
SciCodeCoding#34/ 18052Artificial Analysis2026-09-01
AA-LCRLong context#47/ 17975Artificial Analysis2026-09-01
AA output speedPerformance#78/ 185Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where Qwen3.8 2.4T A95B shows up, it's highlighted.

ReasoningQwen3.8 2.4T A95B · 94
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingQwen3.8 2.4T A95B · 52
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticQwen3.8 2.4T A95B · 82
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextQwen3.8 2.4T A95B · 75
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for Qwen3.8 2.4T A95B in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context75.33%Artificial Analysis2
AA output speedPerformance40 t/sArtificial Analysis2
AA time to first tokenPerformance1.76sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning57.7%Artificial Analysis2
GPQA DiamondReasoning93.5%Artificial Analysis2
Humanity's Last ExamReasoning42.4%Artificial Analysis2
SciCodeCoding51.6%Artificial Analysis2
τ³-Bench BankingAgentic49.07%Artificial Analysis2
Terminal-BenchAgentic82.02%Artificial Analysis2
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.

Reasoning94
Coding52
Agentic82
Long context75
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Qwen3.8 2.4T A95B. Scores link to the original source; gaps mean no public row exists yet.

9 sourced rows

9 of 37 catalog benchmarks have a sourced row for Qwen3.8 2.4T A95B.

Reasoning

Coding

Agentic

Long context

Performance

Changelog
  1. Open weights published. AA Intelligence Index 57.7; hosted list $2 / $6 per 1M tokens.
Sources