AI vs Puzzles

Back to Home
Puzzle#1714

Wordle

Released: 2026-02-27

7
Models Tested
4
Solved
5.57
Avg. Attempts
7
LLM Providers
Correct
Present
Absent

mistralai/mistral-medium-3

mistral

5/6

c
r
a
n
e
d
o
l
l
y
d
u
m
p
y
d
i
a
r
y
d
i
z
z
y

google/gemini-2.0-flash-001

google

Failed

c
r
a
n
e
b
l
i
m
p
v
i
v
i
d
s
i
n
k
d
w
i
g
g
d
f
i
b
e
d

openai/gpt-4o

openai

Failed

c
r
a
n
e
b
l
u
f
f
m
i
d
s
t
d
i
v
v
y
d
i
c
e
y
d
i
a
d
y

anthropic/claude-3.7-sonnet

anthropic

5/6

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

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

meta

Failed

h
o
u
s
e
b
l
a
s
t
f
m
e
n
t
b
r
i
c
k
d
i
g
i
t
d
i
r
g
e

deepseek/deepseek-chat-v3-0324

deepseek

6/6

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

x-ai/grok-3-mini-beta

x-ai

5/6

c
r
a
n
e
g
h
o
s
t
j
u
m
p
y
d
i
l
l
y
d
i
z
z
y

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