Benchmark Heaven
Price & provider filters · adjusted costs
Global
Applies to price views & model offers; benchmark evidence stays unfiltered

Provider explorer

Browse the provider directory on the left — each tagged Hyperscaler / EU / Non-US, with its model count and HQ. Click a provider to see every model it offers and its price; click a model to compare that provider's price against all other providers of the same model, ranked cheapest-first in the selected cost scenario (adjusted USD/task by default).

ProviderFlags# models
OpenRouter · US
HyperscalerEU-capable
15
OpenRouter · US
HyperscalerEU-capable
15
US
HyperscalerEU-capable
14
US
HyperscalerEU-capable
14
Germany
EU-capable
12
OpenRouter · US
11
OpenRouter · US
11
OpenRouter · US
10
OpenRouter · US
9
US
HyperscalerEU-capable
8
OpenRouter · US
HyperscalerEU-capable
8
OpenRouter · US
8
OpenRouter · US
8
Netherlands
EU-capable
8
OpenRouter · US
8
Ireland
EU-capable
8
OpenRouter · US
8
OpenRouter · US
8
US
8
OpenRouter · US
8
OpenRouter · US
8
OpenRouter · US
7
OpenRouter · US
6
OpenRouter · US
6
OpenRouter · US
6
OpenRouter · US
6
OpenRouter · US
5
US
4
OpenRouter · US
4
OpenRouter · US
4
Sweden
EU-capable
4
OpenRouter · Sweden
EU-capable
4
OpenRouter · Spain
EU-capable
3
OpenRouter · US
3
OpenRouter · US
3
OpenRouter · US
3
OpenRouter · US
3
OpenRouter · US
3
OpenRouter · US
3
OpenRouter · Netherlands
EU-capable
2
OpenRouter · US
2
OpenRouter · US
2
OpenRouter · US
2
OpenRouter · US
2
OpenRouter · US
2
OpenRouter · US
2
OpenRouter · France
EU-capable
1
France
EU-capable
1
France
EU-capable
1
OpenRouter · US
1
OpenRouter · US
1
OpenRouter · US
1
OpenRouter ·
1

53 providers. Click one to see its models & prices.

AzureOpenRouterHyperscalerEU-capableHQ: US15 models

Exact OpenRouter provider name. EU-capable only on endpoints explicitly tagged azure/eu or azure/swedencentral; global routes do not inherit EU residency.

Adjusted costs are modeled USD per task: AA output tokens × OpenRouter usage I/O (Chutes global fallback). These are general usage and benchmark proxies for coding-agent work. Missing AA data assumes 1,000 output tokens/task; unknown cache hit assumes 0%; unmeasured additional cache writes assume 0 tokens. Click any underlined price for exact inputs, dates and assumptions. Raw list prices use the selected fixed input/output blend, in USD per million tokens.

ModelComposite (coverage-neutral, dominance-safe percentiles, 0–100) · 5 fixed inputs; Coding Agent v1.4Raw in $/1MRaw out $/1MAdjusted $/taskPrice rankCompare offers
DeepSeek60.0$1.91$3.83#13▸ compare 13
OpenAI65.4$2.50$15.00#2▸ compare 6
Anthropic77.4$5.00$25.00#4▸ compare 8
Anthropic70.8$5.00$25.00#4▸ compare 9
Anthropic60.3$3.00$15.00#4▸ compare 9
OpenAI70.8$0.200$1.20#1 · cheapest▸ compare 6
OpenAI65.4$2.00$12.00#2▸ compare 6
OpenAI77.5$5.00$30.00#3▸ compare 6
OpenAI65.5$5.00$30.00#2▸ compare 6
OpenAI96.0$10.00$50.00#2▸ compare 3
Anthropic95.7$5.00$25.00#5▸ compare 10
Anthropic84.3$5.00$25.00#4▸ compare 9
Anthropic85.8$2.00$10.00#4▸ compare 9
Anthropic93.7$10.00$50.00#4▸ compare 8
Anthropic100.0$10.00$50.00#4▸ compare 7

Raw in/out are unchanged list prices per 1M tokens. The price column follows the active price mode (Adjusted $/task); click any underlined price for exact inputs, sources and assumptions. Click a model row to compare its price across every provider.