Released: 2026-08-26Artificial Analysis

Qwen3.8-Flash-Next

Qwen3.8-Flash-Next from Alibaba, tracked on the Artificial Analysis Intelligence Index.

Intelligence Index
28.6/ 100
Artificial Analysis Intelligence Index (independent composite)
Throughput Speed
77tok/sec
Output streaming throughput
Time to First Answer
87s
Time to first answer token, including any thinking
Cost to Run the Index
0.57USD
Artificial Analysis cost of the full Index run

Measured results · Artificial Analysis

Independent evaluations of Qwen3.8-Flash-Next. Values are reported as published; nothing here is estimated or derived. Snapshot 2026-11-15.

Intelligence Index
28.6
Cost to run the Index
$0.57
Time per task
31.9 min
Omniscience index
-12.2
Reasoning model
Yes
Cached-input price
$0.016 / 1M

Agentic work

Multimodal & science

Full analysis on Artificial Analysis

Headline Evaluation Breakdown

Six independently measured evaluations (Artificial Analysis)
Reasoning · CritPt v2 (Hard)24.5%
Coding & agents · Terminal-Bench 4.025.3%
Long context · AA-LCR79.7%
Knowledge · Humanity's Last Exam38.0%

Sourced Benchmark Evaluations (18)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
CritPtActive11.1%
93
2026-11-15independentSource ↗
AutomationBench-AAActive55.9%
93
2026-11-15independentSource ↗
MLCR-AA (Medical Long Context Reasoning)Active20.6%
93
2026-11-15independentSource ↗
GDP.pdfActive15.6%
93
2026-11-15independentSource ↗
FrontierCode-AA 2.0Active27.1%
93
2026-11-15independentSource ↗
AA-Omniscience (Non-Hallucination)Active53.2%
93
2026-11-15independentSource ↗
AA-Omniscience (Accuracy)Active23.5%
93
2026-11-15independentSource ↗
HLE-DiamondActive16.3%
93
2026-11-15independentSource ↗
CritPt v2 (Hard)Active24.5%
93
2026-11-15independentSource ↗
GPQANearing Saturation92.3%
78
2026-11-15independentSource ↗
AA-LCR (Long Context Reasoning)Active79.7%
93
2026-11-15independentSource ↗
Terminal-Bench-Science 0.1Active7.1%
93
2026-11-15independentSource ↗
Terminal-Bench 4.0Active25.3%
93
2026-11-15independentSource ↗
Terminal-Bench 2.1Active86.1%
93
2026-11-15independentSource ↗
τ³-BankingActive45.4%
93
2026-11-15independentSource ↗
SciCodeActive50.6%
93
2026-11-15independentSource ↗
MMMU-ProActive78.6%
93
2026-11-15independentSource ↗
Humanity's Last ExamActive38%
93
2026-11-15independentSource ↗
Local Hardware Execution (Ollama)Hardware Compatible
# 1. Pull and execute model locally with Ollama
ollama run alibaba-qwen3-8-flash-next

# 2. Or invoke via local OpenAI-compatible endpoint
curl http://localhost:11434/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "alibaba-qwen3-8-flash-next",
    "messages": [{"role": "user", "content": "Analyze reasoning chains on AIME 2026"}]
  }'

Evaluation & Sourcing Notes

Scores listed for Qwen3.8-Flash-Next represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.

All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.