AI vs Puzzles

Back to Home
Puzzle#1699

Wordle

Released: 2026-02-12

8
Models Tested
4
Solved
5.00
Avg. Attempts
8
LLM Providers
Correct
Present
Absent

openai/gpt-4o

openai

2/6

c
r
a
n
e
f
o
r
g
e
d
i
r
g
e
p
u
r
g
e
s
u
r
g
e

mistralai/mistral-medium-3

mistral

Failed

c
r
a
n
e
p
r
o
w
l
d
r
i
e
r
m
e
r
i
t
b
e
r
e
t
c
r
e
m
e

google/gemini-2.0-flash-001

google

Failed

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

anthropic/claude-3.7-sonnet

anthropic

2/6

c
r
a
n
e
p
r
i
d
e
g
r
o
v
e
s
u
r
g
e

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

meta

Failed

h
o
u
s
e
b
l
u
e
s
t
r
u
s
e
b
r
u
s
e
c
r
u
s
e
g
r
u
n
s

deepseek/deepseek-chat-v3-0324

deepseek

Failed

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

x-ai/grok-3-mini-beta

x-ai

2/6

c
r
a
n
e
f
o
r
g
e
m
e
r
g
e
s
u
r
g
e

qwen/qwen3-235b-a22b

qwen

2/6

c
r
a
n
e
t
h
r
e
e
s
u
r
g
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