AI vs Puzzles

Back to Home
Puzzle#1734

Wordle

Released: 2026-03-19

7
Models Tested
2
Solved
5.57
Avg. Attempts
7
LLM Providers
Correct
Present
Absent

google/gemini-2.0-flash-001

google

Failed

r
a
i
n
s
r
o
y
a
l
r
a
v
e
n
r
e
a
c
t
r
e
a
c
h
r
e
a
r
m

openai/gpt-4o

openai

5/6

c
r
a
n
e
r
e
a
p
s
r
e
a
l
m
r
e
a
c
h
r
e
h
a
b

anthropic/claude-3.7-sonnet

anthropic

Failed

s
o
a
r
e
c
r
a
t
e
f
l
a
r
e
b
r
a
k
e
h
e
a
r
d
b
e
r
t
h

deepseek/deepseek-chat-v3-0324

deepseek

Failed

s
l
a
n
t
c
a
m
e
o
b
a
g
e
l
a
b
e
a
m
b
r
e
a
d
b
r
a
k
e

meta-llama/llama-3.3-70b-instruct

meta

Failed

h
o
u
s
e
t
h
e
m
e
h
e
a
v
e
c
h
e
a
t
h
e
a
l
s
h
e
a
r
d

x-ai/grok-3-mini-beta

x-ai

4/6

c
r
a
n
e
e
a
r
t
h
r
h
e
a
s
r
e
h
a
b

mistralai/mistral-medium-3

mistral

Failed

c
r
a
n
e
p
a
r
t
y
m
a
r
r
y
d
a
i
l
y
r
a
z
e
r
r
a
i
s
e

Methodology Note

Each model receives the same prompt structure with Wordle rules and feedback format. Models are given a maximum of 6 attempts with standard positional feedback after each guess to help refine their next attempt.

Leaderboard