Finding 01
The low cost cluster is already deep.
217 priced models sit at or below $0.50 blended per one million tokens using a 3 input to 1 output workload mix. That makes the bottom of the market wide enough for real substitution pressure.
Research note · Live pricing dataset
API Pricing Compression Monitor
A live BenchGecko note reading today's spread of AI API prices: how deep the low cost floor is, which providers anchor it, and which premium models still sit far above the market median.
Data as of
Oct 15, 2026
Model prices · refreshed daily
606
Priced models
62 providers with input and output prices
$0.455
Median input
per one million input tokens
$1.848
Median output
per one million output tokens
217
Low cost band
models at or below $0.50 blended
Finding 02
Premium pricing still has a long tail.
64 priced models sit above $5 blended. The monitor treats these as premium outliers until benchmark strength, latency, context, or specialist capability explains the gap.
Finding 03
Output tokens remain the expensive side.
The median output price is $1.848 versus $0.455 for input. The 90th to 10th percentile spread is 75x for output and 53.6x for input.
Low cost floor
Cheapest blended token prices in the current dataset.
LFM2-2.6B
liquid · input $0.01 · output $0.02
$0.013
blended
LFM2-8B-A1B
liquid · input $0.01 · output $0.02
$0.013
blended
Ling-2.6-flash
inclusionai · input $0.01 · output $0.03
$0.015
blended
Mistral Nemo
Mistral AI · input $0.019 · output $0.03
$0.022
blended
Microsoft Decision 1
Microsoft · input $0.042 · output $0
$0.032
blended
gpt-oss-20b
OpenAI · input $0.018 · output $0.09
$0.036
blended
Granite 4.0 Micro
ibm-granite · input $0.017 · output $0.112
$0.041
blended
Llama 3 8B Lunaris
sao10k · input $0.04 · output $0.05
$0.042
blended
Nex-N2-Mini
nex-agi · input $0.025 · output $0.1
$0.044
blended
Nex-N2.5-Mini
nex-agi · input $0.025 · output $0.1
$0.044
blended
Provider compression table
Providers with at least three priced models, ranked by median blended price.
Premium watchlist
High priced models that need capability context before price alone is interpreted as market power.
o1-pro
OpenAI · input $150 · output $600
$262.5
o1-pro (batch)
OpenAI · input $75 · output $300
$131.25
GPT-5.4 Pro
OpenAI · input $30 · output $180
$67.5
GPT-5.5 Pro
OpenAI · input $30 · output $180
$67.5
Claude Opus 4.6 (Fast)
Anthropic · input $30 · output $150
$60
Claude Opus 4.7 (Fast)
Anthropic · input $30 · output $150
$60
GPT-5.2 Pro
OpenAI · input $21 · output $168
$57.75
GPT-5 Pro
OpenAI · input $15 · output $120
$41.25
Methodology and caveats
Prices are taken from BenchGecko model records and expressed per one million tokens where available.
The blended price uses a 3 input to 1 output workload mix. It is a comparison lens, not a universal workload model.
Compression does not mean quality convergence. Benchmark coverage, latency, context, tools, and reliability still matter.