AI vs Puzzles

Back to Home
Puzzle#1704

Wordle

Released: 2026-02-17

7
Models Tested
0
Solved
6.00
Avg. Attempts
7
LLM Providers
Correct
Present
Absent

openai/gpt-4o

openai

Failed

c
r
a
n
e
p
l
a
i
d
b
a
k
e
d
m
a
d
a
m
g
l
a
d
i
r
a
d
a
r

mistralai/mistral-medium-3

mistral

Failed

c
r
a
n
e
d
a
t
a
l
m
a
d
a
m
b
a
d
d
y
c
a
d
e
t
d
a
f
f
y

google/gemini-2.0-flash-001

google

Failed

c
r
a
n
e
b
l
a
s
t
p
a
w
s
y
m
a
s
o
n
s
a
l
a
d
s
a
f
a
d

anthropic/claude-3.7-sonnet

anthropic

Failed

s
l
a
t
e
s
a
u
n
a
s
p
u
r
s
s
w
u
n
g
s
t
u
c
k
s
h
u
d
h

deepseek/deepseek-chat-v3-0324

deepseek

Failed

c
r
a
n
e
s
a
l
s
a
s
a
p
i
d
s
a
d
l
y
s
w
a
r
d
s
a
n
d
y

x-ai/grok-3-mini-beta

x-ai

Failed

c
r
a
n
e
f
l
a
s
k
b
a
i
t
s
a
s
s
a
y
a
l
i
a
s
v
a
g
a
l

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

meta

Failed

h
o
u
s
e
b
u
r
s
t
l
u
n
c
h
m
u
d
s
y
d
u
c
k
s
d
u
s
k
s

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