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. This snapshot ranks 39 apps by token and request volume. This refresh has no usable model-plus-pricing coverage, so cost estimates are explicitly unavailable rather than inferred.

§ 01 · Findings

Three things the data says.

Snapshot coverage
Volume is available; pricing coverage is not.
The current snapshot has app-level tokens and requests but no usable model-plus-pricing join. Cost and routing-efficiency claims are withheld until that coverage returns.

§ 02 · By cost

Apps ranked by monthly cost.

No usable model-plus-pricing coverage was available in this snapshot. The table remains intentionally empty instead of reporting zero-dollar estimates.

#	App	Categories	Tokens	Requests	Monthly	Blended $/M
Cost breakdown unavailable for snapshot 2026-08-24.

Fig 1 · Cost table withheld because this snapshot has no usable model-plus-pricing coverage.

§ 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	Hermes Agent	#1	41.29T	422.4M
2	Claude Code	#5	10.09T	102.9M
3	Kilo Code	#6	8.08T	98.5M
4	Cline	#7	5.30T	48.3M
5	pi	#9	1.40T	18.0M
6	OpenClaw	#13	139.8B	2.4M
7	Codex	#15	122.6B	2.1M
8	Nous Research API	#17	113.2B	1.9M
9	GitLawb	#126	45.0B	795K
10	Framer	#22	43.2B	345K
11	ISEKAI ZERO	#24	38.2B	1.4M
12	Janitor AI	#25	37.0B	2.3M
13	Fish Audio - the most realistic AI voice generator	#157	35.3B	2.6M
14	Zed Editor	#27	32.8B	215K
15	Sahasra	#28	32.5B	552K
16	Poolside	#184	28.0B	325K
17	Lemonade	#29	27.3B	747K
18	OpenHands	#31	25.9B	254K
19	Descript	#35	21.4B	396K
20	Craft	#37	17.1B	1.1M
21	SillyTavern	#40	13.1B	691K
22	Mira is the leading AI agent inside Telegram	#41	12.6B	847K
23	GDevelop	#42	11.9B	256K
24	Studs.gg	#43	11.8B	326K
25	LangChain	#44	11.4B	822K
26	Roo Code	#50	10.1B	71K
27	Portkey AI	#52	8.4B	216K
28	Qwen Code	#55	7.6B	71K
29	Ito	#58	7.2B	202K
30	Open WebUI	#63	6.6B	182K
31	HammerAI	#68	6.2B	1.7M
32	extra.email	#75	4.5B	526K
33	Chub AI	#78	4.4B	324K
34	OP.GG	#106	2.8B	146K
35	nanobot	#120	2.2B	49K
36	Agent Zero	#123	2.1B	54K
37	Pieces.app: Ambient Artificial Memory	#137	1.8B	205K
38	LibreChat	#187	1.0B	39K
39	novelcrafter	#202	943.2M	25K

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 2026-08-24.
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.