Find Model Alternatives
Select a model to see better, cheaper, or similar alternatives
Data: 2025-03-15 ⢠19 ranked models
Google: Gemini 2.5 Pro
google/gemini-2.5-pro
Arena Score
1470
Rank #1
Price (in/out)
$1.25/$10
per million tokens
Context
1M
tokens
Value Score
261
score per $
đ°Budget Options (Lower Performance)
Mistral: Mistral Small 3
mistralai/mistral-small-24b-instruct-2501
Arena Score
1238-232
Rank
#130
Votes
15.3k
Price (in/out)
$0.049999999999999996/$0.08
vs Selected
86.5x cheaper
Why: Lower performance, much cheaper
High confidence
Qwen: Qwen3 235B A22B Instruct 2507
qwen/qwen3-235b-a22b-2507
Arena Score
1427-43
Rank
#7
Votes
4.2k
Price (in/out)
$0.071/$0.09999999999999999
vs Selected
65.8x cheaper
Why: Lower performance, much cheaper
Medium confidence
Microsoft: Phi 4
microsoft/phi-4
Arena Score
1222-248
Rank
#148
Votes
25.2k
Price (in/out)
$0.06/$0.14
vs Selected
56.3x cheaper
Why: Lower performance, much cheaper
High confidence
Understanding the Data
Arena Score: Higher is better. Based on human preference rankings from LMArena.
Votes:More votes = more reliable score. Models with <1k votes have low confidence.
Value Score: Arena score divided by average price. Higher = better bang for buck.
Confidence:Based on vote count. High = >10k votes, Medium = 1k-10k, Low = <1k.