Codesota · Agentic · OpenRouter modelsInverted OpenRouter viewSnapshot · 2026-06-23
§ Agentic · Model adoption

OpenRouter models,
ranked by real agent usage.

Direct answer: OpenRouter is a model router and catalog. This page ranks OpenRouter models by app adoption, token volume, estimated spend, and top-1 usage so you can choose by coding, price, context length, latency, or breadth instead of reading a flat provider list.

Fig 0 · Concentration

One number explains the whole market.

Of every dollar agents spend on inference, this is who collects it. The top three vendors take more than the rest combined — that's the entire story before any leaderboard.

anthropic
67.2%
$51.08M · 38 apps
openai
12.8%
$9.73M · 35 apps
google
5.1%
$3.86M · 43 apps
minimax
3.8%
$2.86M · 31 apps
deepseek
3.3%
$2.49M · 39 apps
stepfun
1.8%
$1.34M · 7 apps
22 others
6.1%
$4.62M · 193 apps
Of $75.98M total monthly spend, the top three vendors capture 85.1%. Concentration this high is rare — most catalog markets are flatter.
§ 01 · Four winners

One market, four different leaders.

The same 30-day window produces four different #1s depending on what you measure. Dollars, tokens, breadth of adoption, and depth of adoption disagree — that disagreement is the story.

  • Most dollars
    Anthropic: Claude Sonnet 4.6
    $20.90M/month across 29 apps. Premium pricing compounds even at modest volume.
  • Most tokens
    Owl Alpha
    7.95T through 22 apps. High-volume cheap models absorb the long tail.
  • Most adopted
    DeepSeek: DeepSeek V4 Flash
    Shows up in 36 of 46 apps. “Everyone tries it at least a little” is not the same as “anyone uses it primarily”.
  • Most #1 slots
    DeepSeek: DeepSeek V4 Flash
    The top model in 6 of 46 apps. The closest proxy for what agents actually lean on when it matters.
§ 02 · Vendors

Rolled up by vendor.

Same data, one level coarser. Every model per vendor, summed into one row. Blended $/M is vendor-weighted by real usage, so premium vendors show their true price per token.

#VendorModelsAppsTokensMonthly costAvg $/M#1 slots
1anthropic12386.23T$51.08M$8.207
2openai22351.60T$9.73M$6.102
3google14432.54T$3.86M$1.5210
4minimax3315.40T$2.86M$0.534
5deepseek8399.77T$2.49M$0.2611
6stepfun272.89T$1.34M$0.462
7z-ai728867.3B$1.09M$1.260
8xiaomi2272.13T$862K$0.402
9moonshotai420465.1B$662K$1.420
10poolside2122.35T$573K$0.241
11nvidia4171.74T$541K$0.310
12qwen1827339.3B$478K$1.411
13nex-agi111838.8B$386K$0.460
14x-ai379.2B$15K$1.600
15mistralai33222.3B$5K$0.022
16tencent1416.6B$2K$0.100
17aion-labs111.2B$1K$1.020
18sao10k2214.3B$818$0.060
19thedrummer22745.0M$297$0.400
20perplexity1141.2M$262$6.360
21microsoft22382.1M$237$0.620
22anthracite-org1122.4M$80$3.560
23inclusionai122.7B$43$0.020
24nousresearch11137.3M$28$0.210
25gryphe1155.4M$3$0.060
26meta-llama112.9M$0$0.160
27openrouter1227.95T$0—4
28cohere110$0—0
Fig 1 · Vendor rollup. Copper rows mark the top three by monthly cost.
Fig 2 · Cost vs volume

Where every vendor actually sits.

Monthly spend vs token volume, one bubble per vendor. Bubble size scales with apps using that vendor. The diagonal is even pricing — anything above charges more per token than the cluster, anything below charges less.

$0$1$10$100$1K$10K$100K$1.00M$10.00M$100.00M1.0M10.0M100.0M1.0B10.0B100.0B1.00T10.00TTokens / month →↑ Monthly spendanthropicopenaigoogleminimaxdeepseekstepfunz-aixiaomimoonshotaipoolsidenvidiaqwennex-agix-aimistralaitencentaion-labssao10kthedrummerperplexitymicrosoftanthracite-orginclusionainousresearchgryphemeta-llama
Both axes are log-scaled. Bubble area scales with app adoption. Vendors in the upper-left charge more per token; lower-right vendors are the volume-cheap workhorses.
§ 03 · By cost

Top 30 by monthly cost.

Where the dollars actually go. Summed across every app that uses each model.

#ModelVendor$/M in$/M outTokensMonthlyApps#1Top-3
1Anthropic: Claude Sonnet 4.6anthropic$3.00$15.003.29T$20.90M29511
2Anthropic: Claude Opus 4.8anthropic$5.00$25.001.54T$16.32M2414
3OpenAI: GPT-5.5openai$5.00$30.00661.4B$7.94M1613
4Anthropic: Claude Opus 4.7anthropic$5.00$25.00554.2B$5.87M2001
5Anthropic: Claude Opus 4.6anthropic$5.00$25.00536.3B$5.68M2116
6MiniMax: MiniMax M3minimax$0.30$1.204.53T$2.50M2826
7Google: Gemini 3.5 Flashgoogle$1.50$9.00598.6B$2.15M2613
8DeepSeek: DeepSeek V4 Prodeepseek$0.43$0.872.88T$1.60M34311
9StepFun: Step 3.7 Flashstepfun$0.20$1.152.88T$1.34M724
10OpenAI: GPT-5.3-Codexopenai$1.75$14.00246.6B$1.28M512
11Anthropic: Claude Fable 5anthropic$10.00$50.0042.0B$891K600
12Google: Gemini 3 Flash Previewgoogle$0.50$3.00670.3B$804K2137
13DeepSeek: DeepSeek V4 Flashdeepseek$0.09$0.186.42T$739K36617
14Xiaomi: MiMo-V2.5-Proxiaomi$0.43$0.871.27T$708K2613
15MoonshotAI: Kimi K2.6moonshotai$0.66$3.41451.5B$646K1901
16Z.ai: GLM 5.2z-ai$0.95$3.00385.9B$588K1400
17Poolside: Laguna M.1poolside$0.20$0.402.13T$546K1211
18Anthropic: Claude Sonnet 4.5anthropic$3.00$15.0076.6B$487K1301
19Anthropic: Claude Opus 4.8 (Fast)anthropic$10.00$50.0019.1B$405K100
20Qwen: Qwen3.7 Maxqwen$1.25$3.75204.8B$399K1101
21Nex AGI: Nex-N2-Pronex-agi$0.25$1.00838.8B$386K1101
22Z.ai: GLM 5.1z-ai$0.98$3.08239.3B$375K1902
23Google: Gemini 3.1 Pro Preview Custom Toolsgoogle$2.00$12.0073.5B$353K1601
24Anthropic: Claude Haiku 4.5anthropic$1.00$5.00159.6B$338K1102
25OpenAI: GPT-5.4openai$2.50$15.0048.9B$293K1001
26NVIDIA: Nemotron 3 Supernvidia$0.09$0.451.47T$281K1000
27Google: Gemini 3.1 Flash Litegoogle$0.25$1.50435.1B$261K2136
28NVIDIA: Nemotron 3 Ultranvidia$0.50$2.20266.4B$260K1000
29MiniMax: MiniMax M2.7minimax$0.24$0.96571.3B$252K1211
30Anthropic: Claude Opus 4.5anthropic$5.00$25.0015.5B$165K501
§ 04 · By volume

Top 30 by raw token volume.

How much work each model carries. Cheap models lead here.

#ModelVendor$/M in$/M outTokensMonthlyApps#1Top-3
1Owl AlphaFreeopenrouter$0.00$0.007.95T$02249
2DeepSeek: DeepSeek V4 Flashdeepseek$0.09$0.186.42T$739K36617
3MiniMax: MiniMax M3minimax$0.30$1.204.53T$2.50M2826
4Anthropic: Claude Sonnet 4.6anthropic$3.00$15.003.29T$20.90M29511
5DeepSeek: DeepSeek V4 Prodeepseek$0.43$0.872.88T$1.60M34311
6StepFun: Step 3.7 Flashstepfun$0.20$1.152.88T$1.34M724
7Poolside: Laguna M.1poolside$0.20$0.402.13T$546K1211
8Anthropic: Claude Opus 4.8anthropic$5.00$25.001.54T$16.32M2414
9NVIDIA: Nemotron 3 Supernvidia$0.09$0.451.47T$281K1000
10Xiaomi: MiMo-V2.5-Proxiaomi$0.43$0.871.27T$708K2613
11Xiaomi: MiMo-V2.5xiaomi$0.14$0.28857.0B$154K2513
12Nex AGI: Nex-N2-Pronex-agi$0.25$1.00838.8B$386K1101
13Google: Gemini 3 Flash Previewgoogle$0.50$3.00670.3B$804K2137
14OpenAI: GPT-5.5openai$5.00$30.00661.4B$7.94M1613
15Google: Gemini 3.5 Flashgoogle$1.50$9.00598.6B$2.15M2613
16MiniMax: MiniMax M2.7minimax$0.24$0.96571.3B$252K1211
17Anthropic: Claude Opus 4.7anthropic$5.00$25.00554.2B$5.87M2001
18Google: Gemini 2.5 Flash Litegoogle$0.10$0.40537.7B$99K1012
19Anthropic: Claude Opus 4.6anthropic$5.00$25.00536.3B$5.68M2116
20MoonshotAI: Kimi K2.6moonshotai$0.66$3.41451.5B$646K1901
21OpenAI: gpt-oss-120bopenai$0.04$0.18445.6B$35K1401
22Google: Gemini 3.1 Flash Litegoogle$0.25$1.50435.1B$261K2136
23Z.ai: GLM 5.2z-ai$0.95$3.00385.9B$588K1400
24DeepSeek: DeepSeek V3.2deepseek$0.23$0.34322.8B$84K924
25MiniMax: MiniMax M2.5minimax$0.15$0.90290.4B$105K511
26NVIDIA: Nemotron 3 Ultranvidia$0.50$2.20266.4B$260K1000
27OpenAI: GPT-5.3-Codexopenai$1.75$14.00246.6B$1.28M512
28Z.ai: GLM 5.1z-ai$0.98$3.08239.3B$375K1902
29Mistral: Mistral Nemomistralai$0.02$0.03221.6B$5K222
30Poolside: Laguna XS.2poolside$0.10$0.20214.2B$27K200
§ 05 · By breadth

Top 30 by number of apps using it.

Breadth of adoption — models that appear in the top-20 mix of the most apps.

#ModelVendor$/M in$/M outTokensMonthlyApps#1Top-3
1DeepSeek: DeepSeek V4 Flashdeepseek$0.09$0.186.42T$739K36617
2DeepSeek: DeepSeek V4 Prodeepseek$0.43$0.872.88T$1.60M34311
3Anthropic: Claude Sonnet 4.6anthropic$3.00$15.003.29T$20.90M29511
4MiniMax: MiniMax M3minimax$0.30$1.204.53T$2.50M2826
5Xiaomi: MiMo-V2.5-Proxiaomi$0.43$0.871.27T$708K2613
6Google: Gemini 3.5 Flashgoogle$1.50$9.00598.6B$2.15M2613
7Xiaomi: MiMo-V2.5xiaomi$0.14$0.28857.0B$154K2513
8Anthropic: Claude Opus 4.8anthropic$5.00$25.001.54T$16.32M2414
9Owl AlphaFreeopenrouter$0.00$0.007.95T$02249
10Google: Gemini 3 Flash Previewgoogle$0.50$3.00670.3B$804K2137
11Anthropic: Claude Opus 4.6anthropic$5.00$25.00536.3B$5.68M2116
12Google: Gemini 3.1 Flash Litegoogle$0.25$1.50435.1B$261K2136
13Anthropic: Claude Opus 4.7anthropic$5.00$25.00554.2B$5.87M2001
14MoonshotAI: Kimi K2.6moonshotai$0.66$3.41451.5B$646K1901
15Z.ai: GLM 5.1z-ai$0.98$3.08239.3B$375K1902
16OpenAI: GPT-5.5openai$5.00$30.00661.4B$7.94M1613
17Google: Gemini 3.1 Pro Preview Custom Toolsgoogle$2.00$12.0073.5B$353K1601
18Google: Gemini 2.5 Flashgoogle$0.30$2.50157.1B$144K1523
19OpenAI: gpt-oss-120bopenai$0.04$0.18445.6B$35K1401
20Z.ai: GLM 5.2z-ai$0.95$3.00385.9B$588K1400
21Google: Gemma 4 31Bgoogle$0.12$0.3560.2B$11K1400
22Anthropic: Claude Sonnet 4.5anthropic$3.00$15.0076.6B$487K1301
23Poolside: Laguna M.1poolside$0.20$0.402.13T$546K1211
24MiniMax: MiniMax M2.7minimax$0.24$0.96571.3B$252K1211
25Nex AGI: Nex-N2-Pronex-agi$0.25$1.00838.8B$386K1101
26Qwen: Qwen3.7 Maxqwen$1.25$3.75204.8B$399K1101
27Anthropic: Claude Haiku 4.5anthropic$1.00$5.00159.6B$338K1102
28NVIDIA: Nemotron 3 Supernvidia$0.09$0.451.47T$281K1000
29Google: Gemini 2.5 Flash Litegoogle$0.10$0.40537.7B$99K1012
30NVIDIA: Nemotron 3 Ultranvidia$0.50$2.20266.4B$260K1000
§ 06 · By depth

Models that rank #1 in an app's mix (23 models).

Strongest 'primary driver' signal. Being #1 in even one app beats showing up in a long tail.

#ModelVendor$/M in$/M outTokensMonthlyApps#1Top-3
1DeepSeek: DeepSeek V4 Flashdeepseek$0.09$0.186.42T$739K36617
2Anthropic: Claude Sonnet 4.6anthropic$3.00$15.003.29T$20.90M29511
3Owl AlphaFreeopenrouter$0.00$0.007.95T$02249
4DeepSeek: DeepSeek V4 Prodeepseek$0.43$0.872.88T$1.60M34311
5Google: Gemini 3 Flash Previewgoogle$0.50$3.00670.3B$804K2137
6Google: Gemini 3.1 Flash Litegoogle$0.25$1.50435.1B$261K2136
7MiniMax: MiniMax M3minimax$0.30$1.204.53T$2.50M2826
8StepFun: Step 3.7 Flashstepfun$0.20$1.152.88T$1.34M724
9DeepSeek: DeepSeek V3.2deepseek$0.23$0.34322.8B$84K924
10Mistral: Mistral Nemomistralai$0.02$0.03221.6B$5K222
11Google: Gemini 2.5 Flashgoogle$0.30$2.50157.1B$144K1523
12Poolside: Laguna M.1poolside$0.20$0.402.13T$546K1211
13Anthropic: Claude Opus 4.8anthropic$5.00$25.001.54T$16.32M2414
14Xiaomi: MiMo-V2.5-Proxiaomi$0.43$0.871.27T$708K2613
15Xiaomi: MiMo-V2.5xiaomi$0.14$0.28857.0B$154K2513
16OpenAI: GPT-5.5openai$5.00$30.00661.4B$7.94M1613
17Google: Gemini 3.5 Flashgoogle$1.50$9.00598.6B$2.15M2613
18MiniMax: MiniMax M2.7minimax$0.24$0.96571.3B$252K1211
19Google: Gemini 2.5 Flash Litegoogle$0.10$0.40537.7B$99K1012
20Anthropic: Claude Opus 4.6anthropic$5.00$25.00536.3B$5.68M2116
21MiniMax: MiniMax M2.5minimax$0.15$0.90290.4B$105K511
22OpenAI: GPT-5.3-Codexopenai$1.75$14.00246.6B$1.28M512
23Qwen: Qwen3.5-Flashqwen$0.07$0.266.3B$755212
§ 07 · Methodology

How the inverted view is built.

Data source, matching rules, and known limitations — in the open.

  • App leaderboard, inverted.
    Built by inverting the app leaderboard JSON — for every (app, model) pair in the top-20 slice, aggregate by model.
  • Permaslug matching.
    Permaslugs with date suffixes (e.g. anthropic/claude-4.6-opus-20260205) are matched back to the base catalog ID via vendor-scoped token-bag overlap, so dollar totals are not missed.
  • What '#1 in N apps' means.
    The model holds the top token slot for that app's 30-day window. It does not mean the model is objectively best — just that this particular app routes the most tokens to it.
  • Free and stealth lanes.
    Free models contribute to token and adoption rankings but $0 to spend. Stealth / alpha lanes without listed pricing are counted in adoption but not dollars.
  • Regenerate locally.
    python3 scripts/export-openrouter-apps-from-orint.py — rebuilds the snapshot from the cached daily app→model usage. (OpenRouter retired its public /apps/list endpoint in mid-2026, so the live RSC scraper no longer discovers apps on its own.)
§ Final · Related

What to read next.

We reply within 48 hours

Model getting ignored?

Think a model deserves a closer look or isn't priced correctly? Tell us and we'll re-run the match against our catalog.

Tell us what you found →
✓ No newsletter✓ Real humans read this✓ 30 seconds to send