Ir al contenido principal

AI · LEADERBOARD

Elige el modelo que encaja con el trabajo.

Una comparación práctica de los modelos de IA que los equipos pueden desplegar hoy.

Snapshot: 23 models compared

Independent editorial snapshot · not live platform telemetry

Dated editorial snapshot

Review target: weekly and after major releases. Rankings reflect this dated snapshot. OpenRouter usage below has its own source timestamp.

The short answer

Three leaders, three different decisions.

Best overall

Closed model

Anthropic

Intelligence
58.4
Speed
96.6tok/s
Input / output
$4 / $20
Context
1M

Best fit

Demanding reasoning, coding and long-horizon agents

Best open model

Open weights

DeepSeek

Intelligence
32.5
Speed
175tok/s
Input / output
$0.15 / $0.6
Context
1.05M

Best fit

Text and image workloads with a sparse MoE model

Best value

Closed model

Anthropic

Intelligence
36.2
Speed
235.2tok/s
Input / output
$0.1 / $0.5
Context
1M

Best fit

High-volume summarization, subagents and browser use

OpenRouter usage

OpenRouter LLM usage ranking

Top models by prompt, completion, and reasoning tokens processed on OpenRouter. Showing the first 20 only.

Top 20OpenRouter data · cached hourlyUpstream · Nov 17, 2026
  1. 1.DeepSeek V4.1 Flash 1017 (batch)by DeepSeek61.9T tokens15%
  2. 2.Hy4 flashby Tencent22.8T tokensnew
  3. 3.DeepSeek V4.1 Flash (batch)by DeepSeek15.9T tokens6%
  4. 4.Hy4by Tencent14.3T tokens24%
  5. 5.GLM 5.3 Flash (batch)by Z.ai11.5T tokens2%
  6. 6.GPT-6 Luna (batch)by OpenAI8.38T tokens19%
  7. 7.Claude Opus 5.5 (batch)by Anthropic7.53T tokens11%
  8. 8.Step 5 Previewby StepFun6.57T tokens5%
  9. 9.MiMo-V2.6-Flashby Xiaomi4.11T tokens44%
  10. 10.GLM 5.3 (batch)by Z.ai3.62T tokens1%
  11. 11.GPT-6.1 Sol (batch)by OpenAI3.54T tokens13%
  12. 12.Kimi K3.1 Pro (batch)by Moonshot AI3.39T tokens23%
  13. 13.MiniMax M3.1 Flashby MiniMax3.36T tokens22%
  14. 14.DeepSeek V4 Flash 0731 (batch)by DeepSeek3.13T tokens1%
  15. 15.Nemotron 3 Ultra (free)by Nvidia3.12T tokens4%
  16. 16.ERNIE-5.5-Flash-Preview (free)by Baidu2.99T tokensnew
  17. 17.Muse Spark 1.3 Contributorby Meta2.96T tokens2%
  18. 18.Nemotron 3.5 Super VL (free)by Nvidia2.93T tokens2%
  19. 19.Claude Sonnet 5.5 (batch)by Anthropic2.79T tokens17%
  20. 20.Claude Haiku 5.5 (batch)by Anthropic2.77T tokens1%

From benchmark to decision

Which model for which job?

Nine recurring production scenarios, with the trade-offs that matter before a team commits to a model.

Selected use case

01 / 09

Decision view

Customer support

Why it fits

  1. 01235.2 tok/s output
  2. 02$0.10 input / $0.50 output per 1M tokens
  3. 03Positioned for high-volume, cost-sensitive work

How the ranking works

Transparent enough to disagree with.

The ranking combines public benchmark signals with price, measured throughput, context and deployment constraints. It is a decision aid, not a universal definition of model quality.

Dataset snapshot
02 Nov 2026 · 06:45 TST
Review cadence
Revalidated weekly and after major releases

Value score

(intelligence − 25)² × min(1, speed ÷ 30) ÷ (input cost + output cost × 3)

Output is weighted ×3 to reflect output-heavy production use; intelligence is squared so cheap but weak models do not dominate.

Intelligence
Public benchmark and reasoning signals form the default order.
Speed
Output tokens per second; provider and region still affect real latency.
Cost
Public input and output price per million tokens, before enterprise discounts.
Deployment
Context, weight access and hosting constraints determine production fit.

Public references

Prices and performance can change without notice. Validate the finalists against your own prompts, infrastructure and risk requirements before procurement.

FAQ

Questions worth asking before the benchmark wins the argument.

TENTEN AI · MODEL EVALUATION

The leaderboard is the start

Bring the workflow. We’ll test the shortlist.

A model only becomes a good decision after it survives your documents, permissions, latency budget and failure cases.