AI vs Puzzles

Back to Home
Puzzle#1710

Wordle

Released: 2026-02-23

7
Models Tested
4
Solved
5.43
Avg. Attempts
7
LLM Providers
Correct
Present
Absent

mistralai/mistral-medium-3

mistral

Failed

c
r
a
n
e
s
l
a
c
k
c
r
a
c
k
c
a
c
a
o
c
a
b
b
y
c
a
d
d
y

openai/gpt-4o

openai

5/6

s
p
i
l
l
t
r
i
c
e
m
i
t
c
h
o
p
t
i
c
a
t
t
i
c

google/gemini-2.0-flash-001

google

5/6

c
r
a
n
e
b
l
o
c
s
f
i
g
h
t
t
a
i
n
t
a
t
t
i
c

anthropic/claude-3.7-sonnet

anthropic

5/6

s
t
a
r
e
i
t
c
h
y
s
t
o
c
k
w
t
a
c
i
a
t
t
i
c

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

meta

Failed

h
o
u
s
e
b
l
a
s
t
t
r
a
i
n
t
a
i
d
y
f
a
i
t
h
w
a
i
t
t

deepseek/deepseek-chat-v3-0324

deepseek

Failed

c
r
a
n
e
c
l
a
s
p
c
a
b
i
n
a
u
d
i
o
a
m
i
g
o
a
c
r
i
d

x-ai/grok-3-mini-beta

x-ai

5/6

c
r
a
n
e
t
a
c
o
s
c
i
t
a
l
v
a
t
i
c
a
t
t
i
c

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