Released: 2024-09-19

Qwen 2.5 72B Instruct

Alibaba 72B dense open-weight general-purpose instruct model.

Intelligence Index
85.5/ 100
Calibrated multi-domain composite
Throughput Speed
82tok/sec
Output streaming throughput
Time To First Token
340ms
Initial chunk server latency
LiveBench Contam-Free
76.5%
September 2026 suite score

LiveBench Multi-Domain Evaluation Breakdown

Monthly refreshed contamination-free test sets
Logical Reasoning81.0%
Coding & Repo Repair77.0%
Mathematics (AIME)81.5%
Data Analysis & Tables74.0%
Language & Comprehension73.5%
Instruction Following72.0%

Sourced Benchmark Evaluations (2)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
SWE-bench VerifiedSaturated61.6%
25
2024-09-28independentSource ↗
IFEvalDeprecated72%
25
2024-09-23independentSource ↗
Local Hardware Execution (Ollama)Hardware Compatible
# 1. Pull and execute model locally with Ollama
ollama run qwen-2-5-72b-instruct

# 2. Or invoke via local OpenAI-compatible endpoint
curl http://localhost:11434/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-2-5-72b-instruct",
    "messages": [{"role": "user", "content": "Analyze reasoning chains on AIME 2026"}]
  }'

Evaluation & Sourcing Notes

Scores listed for Qwen 2.5 72B Instruct 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.