Codesota · Agentic · OpenRouter appsThe 30-day cost leaderboardSnapshot · 2026-04-27

§ Agentic · Market tracking

Who burns the most on
LLM inference.

Name: OpenRouter App Leaderboard (monthly)
Creator: CodeSOTA

OpenRouter publishes per-app token counts for every AI tool routing through it. We read the public app listings, join each model to live pricing, and rank the top 30 apps by estimated 30-day spend. Cost assumes a 72/28 input/output split — a reasonable approximation for tool-heavy agentic workloads.

§ 01 · Findings

Three things the data says.

Top line item
OpenClaw
$23.54M/month — 31.0% of all spend across the top 30 apps. It moves 15.09T tokens at a blended $1.91/M.
Cheapest routing
HammerAI
Runs the leanest mix at $0.03/M blended — roughly 302.9× cheaper per token than Stanley at $10.52/M. Same job, different model routing.
Tokens ≠ dollars
Model mix dominates.
Claude Code moves 3.76T (more than Hermes Agent's 4.65T) but spends $17.92M vs Hermes's $7.07M. Same category, different checkbook.

§ 02 · By cost

Apps ranked by monthly cost.

Each app's top-20 model mix, priced against OpenRouter's live catalog. A few models (stealth/alpha lanes, unlisted internal routes) carry no price and are excluded from the dollar total but still count toward the token total.

#	App	Categories	Tokens	Requests	Monthly	Blended $/M
1	OpenClaw openclaw.ai		15.09T	267.1M	$23.54M	$1.91
2	Claude Code claude.ai		3.76T	50.2M	$17.92M	$5.20
3	Kilo Code kilocode.ai		6.31T	109.2M	$9.00M	$1.51
4	Hermes Agent hermes-agent.nousresearch.com		4.65T	84.7M	$7.07M	$1.73
5	Descript descript.com		287.0B	5.7M	$6.40M	$5.42
6	Cline cline.bot	programming-app	140.3B	1.8M	$2.36M	$2.26
7	Roo Code github.com/RooVetGit/Roo-Cline		111.8B	1.5M	$1.62M	$3.70
8	LangChain docs.langchain.com		98.9B	6.4M	$1.15M	$5.20
9	VidMuse vidmuse.ai		57.2B	1.8M	$913K	$6.24
10	Ito www.ito.ai	cloud-agent	36.0B	717K	$815K	$5.21
11	SillyTavern sillytavern.app		51.3B	3.2M	$686K	$3.82
12	ISEKAI ZERO www.isekai.world		217.8B	10.6M	$619K	$0.74
13	Agent Zero agent-zero.ai		34.1B	1.6M	$617K	$4.85
14	Lemonade lemonade.gg		118.3B	3.7M	$577K	$1.43
15	Open WebUI openwebui.com	roleplay	33.7B	1.5M	$526K	$5.37
16	Janitor AI janitorai.com		171.3B	14.0M	$519K	$0.77
17	Stanley stan.store		21.2B	733K	$291K	$10.52
18	pi pi.dev		64.3B	1.0M	$251K	$4.31
19	Nous Research API api.nousresearch.com		19.9B	1.2M	$228K	$2.49
20	Chub AI chub.ai		30.1B	2.4M	$198K	$1.85
21	Codex openai.com/codex		28.1B	254K	$167K	$5.96
22	Zed Editor zed.dev	personal-agent	18.7B	321K	$150K	$2.39
23	Gobii gobii.ai		17.8B	379K	$77K	$0.77
24	GDevelop gdevelop.io		20.5B	621K	$73K	$0.88
25	Goose block.github.io/goose		46.4B	2.2M	$72K	$1.65
26	Studs.gg studs.gg		16.6B	508K	$65K	$1.39
27	Mira is the leading consumer AI agent inside Telegram mira.tg		46.9B	1.5M	$47K	$0.50
28	Qwen Code github.com/QwenLM/qwen-code.git	cloud-agent · game	17.5B	285K	$33K	$0.63
29	extra.email extra.email		17.7B	5.6M	$13K	$0.74
30	HammerAI www.hammerai.com	cloud-agent	39.3B	11.5M	$7K	$0.03

Fig 1 · Monthly cost leaderboard. Copper rows are the top three line-items; unpriced stealth/alpha lanes drop from the dollar total but still count toward tokens.

§ 03 · By tokens

Apps ranked by raw token volume.

Full tracked app list including tail entries without model breakdowns. Volume-only ranking — no cost join.

#	App	Global rank	Tokens	Requests
1	OpenClaw	#1	15.09T	267.1M
2	Kilo Code	#2	6.31T	109.2M
3	Hermes Agent	#6	4.65T	84.7M
4	Claude Code	#3	3.76T	50.2M
5	Descript	#8	287.0B	5.7M
6	ISEKAI ZERO	#10	217.8B	10.6M
7	Janitor AI	#11	171.3B	14.0M
8	Cline	#13	140.3B	1.8M
9	Lemonade	#14	118.3B	3.7M
10	Roo Code	#15	111.8B	1.5M
11	LangChain	#16	98.9B	6.4M
12	pi	#17	64.3B	1.0M
13	VidMuse	#18	57.2B	1.8M
14	SillyTavern	#19	51.3B	3.2M
15	Mira is the leading consumer AI agent inside Telegram	#20	46.9B	1.5M
16	Goose	#55	46.4B	2.2M
17	HammerAI	#22	39.3B	11.5M
18	Ito	#24	36.0B	717K
19	Agent Zero	#25	34.1B	1.6M
20	Open WebUI	#26	33.7B	1.5M
21	Chub AI	#28	30.1B	2.4M
22	Codex	#76	28.1B	254K
23	Stanley	#33	21.2B	733K
24	GDevelop	#34	20.5B	621K
25	Nous Research API	#35	19.9B	1.2M
26	Zed Editor	#37	18.7B	321K
27	Gobii	#39	17.8B	379K
28	extra.email	#40	17.7B	5.6M
29	Qwen Code	#41	17.5B	285K
30	Studs.gg	#42	16.6B	508K
31	OpenHands	#43	16.2B	399K
32	Portkey AI	#44	14.8B	1.2M
33	nanobot	#54	11.6B	500K
34	novelcrafter	#82	6.4B	187K
35	LibreChat	#90	5.5B	281K
36	ZeroClaw	#99	4.5B	160K
37	CoffeeCat AI Image Generator	#103	4.3B	1.7M
38	openbench	#112	3.7B	654K
39	Fish Audio - the most realistic AI voice generator	#151	2.6B	379K
40	Nexu Link	#154	2.6B	107K
41	Aider	#166	2.4B	73K

Fig 2 · All tracked apps ranked by tokens. The sparkline on row 1 shows the top-10 volume curve — long-tailed, as expected from a power-law market.

§ 04 · Methodology

Where these numbers come from.

Data source.
openrouter.ai/apps and each app's detail page, fetched via the React Server Components stream (RSC: 1 header). Captured 2026-04-27.
Pricing.
Public openrouter.ai/api/v1/models catalog. Permaslugs (e.g. anthropic/claude-4.6-opus-20260205) are matched back to base catalog IDs after stripping date and preview suffixes.
Blended $/M.
price_in × 0.72 + price_out × 0.28. Agentic workloads are context-heavy because every tool round replays the conversation. A 50/50 split inflates dollar totals by ~10–15%.
Top-20 models per app.
OpenRouter surfaces the top 20 models by tokens for each app. Apps may use more models beyond that slice at long-tail volume, so reported spend is a lower bound.
Unpriced models.
Stealth/alpha lanes (e.g. openrouter/hunter-alpha) and unlisted providers are dropped from the dollar total. Free tiers are priced at $0.
Not a quality ranking.
This page measures compute spend, not output quality. For task performance see our agentic benchmarks.

§ Final · Related

What to read next.

We reply within 48 hours

App missing from the leaderboard?

Know an app routing serious volume through OpenRouter that isn't here? Send the slug — we'll include it in the next refresh.

Tell us what you found →

✓ No newsletter✓ Real humans read this✓ 30 seconds to send

Who burns the most on LLM inference.