JOHNCLICK.AI
HOST DIAGNOSTIC · Alibaba

Qwen3 Max Instruct

Qwen3 Max Instruct · ACTIVE· Qwen· CLOSED
ACTIVE
ATTRIBUTE MATRIX · ATRBT GROUP 01
Bulk Apperception
[14]
Reasoning
[14]
Math Precision
[14]
Code Generation
[13]
World Knowledge
[15]
Scientific Acumen
[14]
Instruction Following
[14]
Context Fidelity
[14]
Candor
[15]
Creativity
[13]
Multi-turn Coherence
[14]
Multimodal Fluency
[13]
Tool Use
[13]
Planning
[13]
Self-Correction
[13]
Tenacity
[13]
Safety Alignment
[13]
Calibration
[14]
Speed
[15]
Cost Efficiency
[17]
HOST TELEMETRY
AVG
13.9
/20
BEST
17
Cost Efficiency
WEAK
13
Code Generation
PARAMETERS
Undisclosed
CONTEXT
—
PRICING
—
STATUS
ACTIVE
ATTRIBUTE SCORES · 20 DIMENSIONS
Cognitive
Bulk Apperce…
14
Reasoning
14
Mathematical…
14
World Knowle…
15
Scientific A…
14
Technical
Code Generat…
13
Tool Use
13
Multimodal F…
13
Speed
15
Cost Efficie…
17
Behavioral
Candor
15
Creativity
13
Tenacity
13
Self-Correct…
13
Calibration
14
Operational
Instruction …
14
Context Fide…
14
Multi-turn C…
14
Planning and…
13
Safety Align…
13
SCORING METHODOLOGY
CALIBRATED per ADR-NM108 (April 2026 snapshot). SimpleQA Verified: 67% (was #1 before Gemini 3 Pro took 73%). Candor scored high (15) reflecting SimpleQA leadership. Qwen3 generation. On LM Arena WebDev. May have data contamination concerns per Epoch AI (ranks lower on AA-Omniscience, a closed benchmark similar to SimpleQA). Cost 17 (open weights, Apache 2.0).
← ALL HOSTS
COMPARE IN MATRIX →KNOWLEDGE GRAPH →