AI vs Puzzles

Back to Home
Puzzle#1698

Wordle

Released: 2026-02-11

8
Models Tested
1
Solved
6.00
Avg. Attempts
8
LLM Providers
Correct
Present
Absent

mistralai/mistral-medium-3

mistral

Failed

c
r
a
n
e
m
a
n
e
t
l
a
n
a
i
d
a
n
c
e
b
l
a
n
d
a
m
a
z
e

openai/gpt-4o

openai

Failed

s
h
a
d
e
g
a
b
l
e
e
a
g
e
r
m
a
n
g
a
a
n
g
e
l
g
l
e
a
n

anthropic/claude-3.7-sonnet

anthropic

Failed

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

google/gemini-2.0-flash-001

google

Failed

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

deepseek/deepseek-chat-v3-0324

deepseek

Failed

c
r
a
n
e
a
n
o
d
e
e
n
a
c
t
e
a
t
e
n
a
m
e
n
d
a
n
e
n
t

x-ai/grok-3-mini-beta

x-ai

Failed

c
r
a
n
e
w
a
n
e
s
n
e
p
a
l
k
e
n
a
f
b
e
a
n
s
h
e
m
a
n

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

meta

Failed

h
o
u
s
e
f
l
a
m
e
s
p
a
c
e
t
a
s
t
e
b
r
a
n
e
g
a
n
e
y

qwen/qwen3-235b-a22b

qwen

2/6

c
r
a
n
e
a
l
i
e
n
e
a
t
e
n
a
a
r
o
n
b
e
v
a
n
v
e
g
a
n

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