Budget · Under $5/M
LLMs under $5 per million tokens
Every model priced below $5 per 1M input tokens. Ranked by benchmark score within budget.
Models40
Top quality82.2
Max price$5.00/M
What this page is
This page ranks every model with input priced below $5 per million tokens by benchmark score. The sub-$5 tier is the pragmatic production sweet spot · it catches strong mid-frontier models (Sonnet, GPT-5 mini, Gemini 2.5 Pro) as well as every open-source option. For most real-world apps, the top entry here will match Opus or GPT-5 on 90% of tasks at a fraction of the cost.
Ranked by quality within budget
Input under $5/M, highest benchmark score first.
| # | Model | In $/1M | Out $/1M | Type |
|---|---|---|---|---|
| 1 | $4.00 | $20.00 | Closed | |
| 2 | $1.25 | $10.00 | Closed | |
| 3 | $0.40 | $1.20 | OSS | |
| 4 | $1.25 | $10.00 | Closed | |
| 5 | $0.10 | $0.30 | OSS | |
| 6 | $2.00 | $10.00 | Closed | |
| 7 | $0.09 | $0.29 | OSS | |
| 8 | $1.25 | $10.00 | Closed | |
| 9 | $1.10 | $4.40 | Closed | |
| 10 | $1.75 | $14.00 | Closed | |
| 11 | $0.60 | $6.00 | Closed | |
| 12 | $3.00 | $15.00 | Closed | |
| 13 | $0.30 | $1.20 | OSS | |
| 14 | $0.60 | $2.20 | OSS | |
| 15 | $1.00 | $2.00 | Closed | |
| 16 | $0.00 | $0.00 | OSS | |
| 17 | $1.25 | $10.00 | Closed | |
| 18 | $2.00 | $12.00 | Closed | |
| 19 | $2.00 | $10.00 | Closed | |
| 20 | $0.00 | $0.00 | OSS | |
| 21 | $0.20 | $2.40 | OSS | |
| 22 | $0.30 | $0.50 | Closed | |
| 23 | $0.07 | $0.30 | Closed | |
| 24 | $0.03 | $1.28 | OSS | |
| 25 | $0.15 | $1.20 | OSS | |
| 26 | $1.25 | $3.75 | OSS | |
| 27 | $0.25 | $2.00 | Closed | |
| 28 | $1.10 | $4.40 | Closed | |
| 29 | $0.55 | $5.00 | Closed | |
| 30 | $0.55 | $3.50 | OSS | |
| 31 | $0.20 | $0.80 | OSS | |
| 32 | $0.15 | $0.60 | Closed | |
| 33 | $0.67 | $14.00 | Closed | |
| 34 | $1.00 | $3.00 | Closed | |
| 35 | $0.50 | $2.15 | OSS | |
| 36 | $0.54 | $0.54 | OSS | |
| 37 | $0.20 | $1.20 | Closed | |
| 38 | $0.75 | $3.75 | Closed | |
| 39 | $2.00 | $6.00 | Closed | |
| 40 | $0.75 | $3.75 | Closed |
Top 3 best-quality sub-$5 models
Best quality under $5
Claude Opus 5.5
input
$4.00/M
output
$20.00/M
Claude Opus 5.5 scores 82.2 on our composite while pricing below $5/M input. Strong default for production workloads with budget constraints.
Runner up
GPT-5 Chat
input
$1.25/M
output
$10.00/M
GPT-5 Chat scores 81.9 on our composite while pricing below $5/M input. Strong default for production workloads with budget constraints.
Third
DeepSeek V3.2 Speciale
input
$0.40/M
output
$1.20/M
DeepSeek V3.2 Speciale scores 78.2 on our composite while pricing below $5/M input. Strong default for production workloads with budget constraints.
The price gap · cheapest vs most expensive
Cheapest
Claude Opus 5.5
$4.00/M
$ per 1M input tokens
Why the gap
Within this tier, price differences are small but quality varies a lot. The top entry (ranked by avg score) often beats several "premium" models that sit above $5.
Most expensive
Claude Opus 5.5
$4.00/M
$ per 1M input tokens
Frequently asked questions
Mid-tier. Frontier models run $10 to $75 per million. Bulk open-source runs under $1. $5 catches capable mid-size models (Sonnet at some providers, GPT-5 mini, Gemini 2.5 Pro at scale).