JOHNCLICK.AI
HOST DIAGNOSTIC · Anthropic

Claude 3.7 Sonnet

Claude 3.7 Sonnet · ACTIVE· Claude· CLOSED
ACTIVE
ATTRIBUTE MATRIX · ATRBT GROUP 01
Bulk Apperception
[11]
Reasoning
[11]
Math Precision
[9]
Code Generation
[13]
World Knowledge
[11]
Scientific Acumen
[10]
Instruction Following
[12]
Context Fidelity
[11]
Candor
[11]
Creativity
[10]
Multi-turn Coherence
[11]
Multimodal Fluency
[8]
Tool Use
[10]
Planning
[10]
Self-Correction
[10]
Tenacity
[10]
Safety Alignment
[13]
Calibration
[11]
Speed
[12]
Cost Efficiency
[12]
HOST TELEMETRY
AVG
10.8
/20
BEST
13
Code Generation
WEAK
8
Multimodal Fluency
PARAMETERS
Undisclosed
CONTEXT
—
PRICING
—
STATUS
ACTIVE
ATTRIBUTE SCORES · 20 DIMENSIONS
Cognitive
Bulk Apperce…
11
Reasoning
11
Mathematical…
9
World Knowle…
11
Scientific A…
10
Technical
Code Generat…
13
Tool Use
10
Multimodal F…
8
Speed
12
Cost Efficie…
12
Behavioral
Candor
11
Creativity
10
Tenacity
10
Self-Correct…
10
Calibration
11
Operational
Instruction …
12
Context Fide…
11
Multi-turn C…
11
Planning and…
10
Safety Align…
13
SCORING METHODOLOGY
CALIBRATED per ADR-NM108 (April 2026 snapshot). Bridge between 3.5 Sonnet and Sonnet 4. Referenced in Epoch AI + LM Council benchmarks. SWE-Bench ~49%. Introduced early extended thinking capability. Arena Text: ~1350-1380 est → 12-14 band, scored conservatively at 11 as a transitional model.
← ALL HOSTS
COMPARE IN MATRIX →KNOWLEDGE GRAPH →