Data updated November 21, 2026

Agents that use tools

Multi-step work where the model calls tools on its own.

Models that have a measurement or a declared feature for this task, with the evidence and the cost of a standard job. The size of the job is our assumption.

Models listed
64
Measure used for the order
Agentic Index

List entries: 65. A free route of a model is its own entry, next to the paid model.

Order: By score By cost

Our assumed job. Requests: 200. Tokens per request: 45,000 in, 2,500 out.

Cost of this job. Our assumption for this job. Requests: 200. Input tokens per request: 45,000. Output tokens per request: 2,500. Share of input served from cache: 75%. The prices are real; the size of the job is not. Thinking is billed as output tokens; a model that thinks longer costs more than this estimate.

Rank among scored models. A band label appears only where at least 30 models are scored. Ranks are among all models we list, including older ones. The share is rounded up to whole models, so a rank slightly past the exact share still counts. Models with equal scores share the better rank.

No free model reaches the top 25% here.

Free route. Rate limits apply and the route can differ from the paid one.

Free Agents that use tools: Gemma 4 26B A4B · Rank 44 of 64, among scored models.

Top 10% 7

  1. Claude Fable 5

    Anthropic · On OpenRouter since June 9, 2026

    Artificial AnalysisAgentic Index: 50.7 · Rank 1 of 64, among scored models

    No cheaper model scores as high

    $54.64 Cost of the job $20.00 · mixed / 1M tokens
  2. Qwen3.8 2.4T A95B

    Qwen · On OpenRouter since August 12, 2026

    Artificial AnalysisAgentic Index: 50.1 · Rank 2 of 64, among scored models

    No cheaper model scores as high

    $9.26 Cost of the job $3.00 · mixed / 1M tokens
  3. DeepSeek V4 Flash Vision Exp

    DeepSeek · On OpenRouter since August 21, 2026

    Artificial AnalysisAgentic Index: 47.5 · Rank 3 of 64, among scored models

    No cheaper model scores as high

    $1.76 Cost of the job $0.66 · mixed / 1M tokens via GMICloud
  4. GPT-5.6 Sol

    OpenAI · On OpenRouter since July 9, 2026

    Artificial AnalysisAgentic Index: 47.4 · Rank 4 of 64, among scored models

    Models that score at least as high for less in this scenario: 2

    $21.86 Cost of the job $8.00 · mixed / 1M tokens Price rises above 272k tokens
  5. Claude Sonnet 5

    Anthropic · On OpenRouter since June 30, 2026

    Artificial AnalysisAgentic Index: 43.6 · Rank 5 of 64, among scored models

    Models that score at least as high for less in this scenario: 2

    $10.93 Cost of the job $4.00 · mixed / 1M tokens
  6. Muse Spark 1.2

    Meta · On OpenRouter since August 5, 2026

    Artificial AnalysisAgentic Index: 43.2 · Rank 6 of 64, among scored models

    Models that score at least as high for less in this scenario: 1

    $5.99 Cost of the job $2.00 · mixed / 1M tokens
  7. Claude Opus 4.8

    Anthropic · On OpenRouter since May 27, 2026

    Artificial AnalysisAgentic Index: 41.9 · Rank 7 of 64, among scored models

    Models that score at least as high for less in this scenario: 5

    $27.32 Cost of the job $10.00 · mixed / 1M tokens

Top 25% 9

  1. Grok 4.5

    SpaceXSI · On OpenRouter since July 8, 2026

    Artificial AnalysisAgentic Index: 41.2 · Rank 8 of 64, among scored models

    Models that score at least as high for less in this scenario: 3

    $9.58 Cost of the job $3.00 · mixed / 1M tokens Price rises above 200k tokens
  2. DeepSeek V4 Flash 0731

    DeepSeek · On OpenRouter since July 31, 2026

    Artificial AnalysisAgentic Index: 41.0 · Rank 9 of 64, among scored models

    $0.09 mixed / 1M tokens in $0.06 · out $0.18 via DeepInfra Price disputed
  3. Claude Opus 4.7

    Anthropic · On OpenRouter since April 16, 2026

    Artificial AnalysisAgentic Index: 38.6 · Rank 10 of 64, among scored models

    Models that score at least as high for less in this scenario: 6

    $27.32 Cost of the job $10.00 · mixed / 1M tokens
  4. GLM 5.2

    Z.ai · On OpenRouter since June 16, 2026

    Artificial AnalysisAgentic Index: 38.4 · Rank 11 of 64, among scored models

    $2.15 mixed / 1M tokens in $1.40 · out $4.40 Price disputed
  5. GPT-5.5

    OpenAI · On OpenRouter since April 24, 2026

    Artificial AnalysisAgentic Index: 36.4 · Rank 12 of 64, among scored models

    Models that score at least as high for less in this scenario: 8

    $29.78 Cost of the job $11.25 · mixed / 1M tokens Price rises above 272k tokens
  6. Gemini 3.7 Flash

    Google · On OpenRouter since August 13, 2026

    Artificial AnalysisAgentic Index: 35.2 · Rank 13 of 64, among scored models

    Models that score at least as high for less in this scenario: 1

    $4.09 Cost of the job $1.50 · mixed / 1M tokens promo
  7. Claude Sonnet 4.6

    Anthropic · On OpenRouter since February 17, 2026

    Artificial AnalysisAgentic Index: 31.8 · Rank 14 of 64, among scored models

    Models that score at least as high for less in this scenario: 6

    $16.39 Cost of the job $6.00 · mixed / 1M tokens
  8. Gemini 3.6 Flash

    Google · On OpenRouter since July 21, 2026

    Artificial AnalysisAgentic Index: 29.0 · Rank 15 of 64, among scored models

    Models that score at least as high for less in this scenario: 1

    $4.09 Cost of the job $1.50 · mixed / 1M tokens
  9. Ling 3.0 Flash VL

    inclusionAI · On OpenRouter since September 10, 2026

    Artificial AnalysisAgentic Index: 28.7 · Rank 16 of 64, among scored models

    $0.31 Cost of the job $0.09 · mixed / 1M tokens via DeepInfra

Scored, outside the top 25% 48

  1. Motif 3

    Motif Technologies · On OpenRouter since November 2, 2026

    Artificial AnalysisAgentic Index: 28.4 · Rank 17 of 64, among scored models

    Models that score at least as high for less in this scenario: 1

    $1.69 Cost of the job $0.525 · mixed / 1M tokens via KT
  2. Gemini 3.5 Flash

    Google · On OpenRouter since May 19, 2026

    Artificial AnalysisAgentic Index: 26.0 · Rank 18 of 64, among scored models

    Models that score at least as high for less in this scenario: 6

    $8.93 Cost of the job $3.38 · mixed / 1M tokens
  3. Muse Spark 1.1

    Meta · On OpenRouter since July 16, 2026

    Artificial AnalysisAgentic Index: 25.8 · Rank 19 of 64, among scored models

    Models that score at least as high for less in this scenario: 5

    $5.99 Cost of the job $2.00 · mixed / 1M tokens
  4. Hy3 preview

    Tencent · On OpenRouter since April 22, 2026

    Artificial AnalysisAgentic Index: 24.1 · Rank 20 of 64, among scored models

    Models that score at least as high for less in this scenario: 1

    $1.11 Cost of the job $0.285 · mixed / 1M tokens via GMICloud
  5. GLM 5.1

    Z.ai · On OpenRouter since April 7, 2026

    Artificial AnalysisAgentic Index: 23.9 · Rank 21 of 64, among scored models

    Models that score at least as high for less in this scenario: 8

    $7.14 Cost of the job $2.15 · mixed / 1M tokens
  6. Qwen3.7 Max

    Qwen · On OpenRouter since May 21, 2026

    Artificial AnalysisAgentic Index: 22.5 · Rank 22 of 64, among scored models

    Models that score at least as high for less in this scenario: 9

    $7.57 Cost of the job $2.21 · mixed / 1M tokens
  7. DeepSeek V4 Flash 0423

    DeepSeek · On OpenRouter since April 24, 2026

    Artificial AnalysisAgentic Index: 22.2 · Rank 23 of 64, among scored models

    $0.1125 mixed / 1M tokens in $0.09 · out $0.18 via DeepInfra Price disputed
  8. MiMo-V2.5-Pro

    Xiaomi · On OpenRouter since April 22, 2026

    Artificial AnalysisAgentic Index: 21.3 · Rank 24 of 64, among scored models

    Models that score at least as high for less in this scenario: 4

    $2.64 Cost of the job $0.81 · mixed / 1M tokens via DigitalOcean
  9. Kimi K2.7 Code

    MoonshotAI · On OpenRouter since June 12, 2026

    Artificial AnalysisAgentic Index: 21.0 · Rank 25 of 64, among scored models

    Models that score at least as high for less in this scenario: 7

    $5.45 Cost of the job $1.71 · mixed / 1M tokens
  10. Kimi K2.6

    MoonshotAI · On OpenRouter since April 20, 2026

    Artificial AnalysisAgentic Index: 20.5 · Rank 26 of 64, among scored models

    Models that score at least as high for less in this scenario: 7

    $5.24 Cost of the job $1.71 · mixed / 1M tokens
  11. Ling 3.0 Flash

    inclusionAI · On OpenRouter since July 23, 2026

    Artificial AnalysisAgentic Index: 19.3 · Rank 27 of 64, among scored models

    $0.11 Cost of the job $0.0315 · mixed / 1M tokens via Novita promo
  12. Qwen3.6 27B

    Qwen · On OpenRouter since April 27, 2026

    Artificial AnalysisAgentic Index: 18.5 · Rank 28 of 64, among scored models

    Models that score at least as high for less in this scenario: 9

    $5.40 Cost of the job $1.01 · mixed / 1M tokens
  13. GPT-5.4 Mini

    OpenAI · On OpenRouter since March 17, 2026

    Artificial AnalysisAgentic Index: 17.9 · Rank 29 of 64, among scored models

    Models that score at least as high for less in this scenario: 8

    $4.47 Cost of the job $1.69 · mixed / 1M tokens
  14. Qwen3.7 Plus

    Qwen · On OpenRouter since June 3, 2026

    Artificial AnalysisAgentic Index: 17.5 · Rank 30 of 64, among scored models

    Models that score at least as high for less in this scenario: 5

    $1.80 Cost of the job $0.56 · mixed / 1M tokens Price rises above 256k tokens
  15. GPT-5.4 Nano

    OpenAI · On OpenRouter since March 17, 2026

    Artificial AnalysisAgentic Index: 16.0 · Rank 31 of 64, among scored models

    Models that score at least as high for less in this scenario: 3

    $1.22 Cost of the job $0.4625 · mixed / 1M tokens
  16. MiMo-V2.5

    Xiaomi · On OpenRouter since April 22, 2026

    Artificial AnalysisAgentic Index: 15.8 · Rank 32 of 64, among scored models

    Models that score at least as high for less in this scenario: 7

    $2.45 Cost of the job $0.80 · mixed / 1M tokens via Venice
  17. Claude Sonnet 4.5

    Anthropic · On OpenRouter since September 29, 2025

    Artificial AnalysisAgentic Index: 15.8 · Rank 32 of 64, among scored models

    Models that score at least as high for less in this scenario: 23

    $16.39 Cost of the job $6.00 · mixed / 1M tokens Price rises above 200k tokens
  18. Grok 4.3

    SpaceXSI · On OpenRouter since April 30, 2026

    Artificial AnalysisAgentic Index: 15.5 · Rank 34 of 64, among scored models

    Models that score at least as high for less in this scenario: 15

    $5.45 Cost of the job $1.56 · mixed / 1M tokens Price rises above 200k tokens
  19. MiniMax M2.7

    MiniMax · On OpenRouter since March 18, 2026

    Artificial AnalysisAgentic Index: 15.3 · Rank 35 of 64, among scored models

    Models that score at least as high for less in this scenario: 18

    $6.30 Cost of the job $0.90 · mixed / 1M tokens via Groq
  20. LongCat 2.0

    Meituan · On OpenRouter since July 20, 2026

    Artificial AnalysisAgentic Index: 14.0 · Rank 36 of 64, among scored models

    Models that score at least as high for less in this scenario: 4

    $1.33 Cost of the job $0.525 · mixed / 1M tokens via AtlasCloud promo
  21. Qwen3.6 35B A3B

    Qwen · On OpenRouter since April 27, 2026

    Artificial AnalysisAgentic Index: 13.1 · Rank 37 of 64, among scored models

    Models that score at least as high for less in this scenario: 2

    $0.91 Cost of the job $0.2975 · mixed / 1M tokens via CoreWeave
  22. Qwen3.5 397B A17B

    Qwen · On OpenRouter since February 16, 2026

    Artificial AnalysisAgentic Index: 8.3 · Rank 38 of 64, among scored models

    Models that score at least as high for less in this scenario: 14

    $4.68 Cost of the job $0.8775 · mixed / 1M tokens
  23. Gemini 3.1 Pro Preview

    Google · On OpenRouter since February 19, 2026

    Artificial AnalysisAgentic Index: 8.2 · Rank 39 of 64, among scored models

    Models that score at least as high for less in this scenario: 28

    $11.91 Cost of the job $4.50 · mixed / 1M tokens Price rises above 200k tokens
  24. Qwen3.5-122B-A10B

    Qwen · On OpenRouter since February 25, 2026

    Artificial AnalysisAgentic Index: 7.6 · Rank 40 of 64, among scored models

    Models that score at least as high for less in this scenario: 11

    $3.38 Cost of the job $0.715 · mixed / 1M tokens
  25. GPT-5 Mini

    OpenAI · On OpenRouter since August 7, 2025

    Artificial AnalysisAgentic Index: 6.8 · Rank 41 of 64, among scored models

    Ends December 11, 2026

    Models that score at least as high for less in this scenario: 7

    $1.74 Cost of the job $0.6875 · mixed / 1M tokens
  26. DeepSeek V3.1 Terminus

    DeepSeek · On OpenRouter since September 22, 2025

    Artificial AnalysisAgentic Index: 6.8 · Rank 41 of 64, among scored models

    Models that score at least as high for less in this scenario: 12

    $2.93 Cost of the job $0.4525 · mixed / 1M tokens via SiliconFlow
  27. Ling 3.0 Tiny

    inclusionAI · On OpenRouter since August 6, 2026

    Artificial AnalysisAgentic Index: 4.4 · Rank 43 of 64, among scored models

    $0.08 Cost of the job $0.0225 · mixed / 1M tokens via Ant Digital
  28. Gemma 4 26B A4B

    Google · On OpenRouter since April 3, 2026

    Artificial AnalysisAgentic Index: 2.9 · Rank 44 of 64, among scored models

    Free version

    The free route differs in: Max output: 33k (other version: 236k); Structured output: Not listed (other version: Yes).

    $0.2625 mixed / 1M tokens in $0.15 · out $0.60 Price disputed
  29. Devstral 2 2512

    Mistral · On OpenRouter since December 9, 2025

    Artificial AnalysisAgentic Index: 2.3 · Rank 45 of 64, among scored models

    Models that score at least as high for less in this scenario: 11

    $2.18 Cost of the job $0.80 · mixed / 1M tokens
  30. Mistral Medium 3.1

    Mistral · On OpenRouter since August 13, 2025

    Artificial AnalysisAgentic Index: 1.9 · Rank 46 of 64, among scored models

    Models that score at least as high for less in this scenario: 12

    $2.40 Cost of the job $0.88 · mixed / 1M tokens
  31. Nemotron 3 Super

    NVIDIA · On OpenRouter since March 11, 2026

    Artificial AnalysisAgentic Index: 1.7 · Rank 47 of 64, among scored models

    Models that score at least as high for less in this scenario: 3

    $0.88 Cost of the job $0.1563 · mixed / 1M tokens via DeepInfra
  32. Gemini 2.5 Pro

    Google · On OpenRouter since June 17, 2025

    Artificial AnalysisAgentic Index: 1.6 · Rank 48 of 64, among scored models

    Models that score at least as high for less in this scenario: 31

    $8.69 Cost of the job $3.44 · mixed / 1M tokens Price rises above 200k tokens
  33. Qwen3 235B A22B Thinking 2507

    Qwen · On OpenRouter since July 25, 2025

    Artificial AnalysisAgentic Index: 1.3 · Rank 49 of 64, among scored models

    Models that score at least as high for less in this scenario: 26

    $5.80 Cost of the job $1.21 · mixed / 1M tokens via Venice
  34. gpt-oss-20b

    OpenAI · On OpenRouter since August 5, 2025

    Artificial AnalysisAgentic Index: 1.2 · Rank 50 of 64, among scored models

    Models that score at least as high for less in this scenario: 2

    $0.14 Cost of the job $0.034 · mixed / 1M tokens via Darkbloom
  35. Qwen3.5-9B

    Qwen · On OpenRouter since March 10, 2026

    Artificial AnalysisAgentic Index: 1.2 · Rank 50 of 64, among scored models

    Models that score at least as high for less in this scenario: 4

    $0.83 Cost of the job $0.1018 · mixed / 1M tokens via PrimeIntellect
  36. Ministral 3 14B 2512

    Mistral · On OpenRouter since December 2, 2025

    Artificial AnalysisAgentic Index: 1.1 · Rank 52 of 64, among scored models

    Models that score at least as high for less in this scenario: 4

    $0.69 Cost of the job $0.20 · mixed / 1M tokens
  37. Nemotron 3 Nano 30B A3B

    NVIDIA · On OpenRouter since December 14, 2025

    Artificial AnalysisAgentic Index: 1.0 · Rank 53 of 64, among scored models

    Models that score at least as high for less in this scenario: 4

    $0.34 Cost of the job $0.105 · mixed / 1M tokens via Nebius
  38. Mistral Large 3 2512

    Mistral · On OpenRouter since December 1, 2025

    Artificial AnalysisAgentic Index: 1.0 · Rank 53 of 64, among scored models

    Models that score at least as high for less in this scenario: 17

    $2.23 Cost of the job $0.75 · mixed / 1M tokens
  39. Qwen3 Coder Next

    Qwen · On OpenRouter since February 4, 2026

    Artificial AnalysisAgentic Index: 0.9 · Rank 55 of 64, among scored models

    Models that score at least as high for less in this scenario: 10

    $1.14 Cost of the job $0.29 · mixed / 1M tokens via Parasail
  40. Qwen3 14B

    Qwen · On OpenRouter since April 28, 2025

    Artificial AnalysisAgentic Index: 0.9 · Rank 55 of 64, among scored models

    $0.15 mixed / 1M tokens in $0.12 · out $0.24 via DeepInfra
  41. Qwen3 32B

    Qwen · On OpenRouter since April 28, 2025

    Artificial AnalysisAgentic Index: 0.9 · Rank 55 of 64, among scored models

    $0.13 mixed / 1M tokens in $0.08 · out $0.28 via DeepInfra
  42. Ministral 3 3B 2512

    Mistral · On OpenRouter since December 2, 2025

    Artificial AnalysisAgentic Index: 0.8 · Rank 58 of 64, among scored models

    Models that score at least as high for less in this scenario: 5

    $0.35 Cost of the job $0.10 · mixed / 1M tokens
  43. DeepSeek V3 0324

    DeepSeek · On OpenRouter since March 24, 2025

    Artificial AnalysisAgentic Index: 0.8 · Rank 58 of 64, among scored models

    Models that score at least as high for less in this scenario: 23

    $2.75 Cost of the job $0.4375 · mixed / 1M tokens via SiliconFlow
  44. Ministral 3 8B 2512

    Mistral · On OpenRouter since December 2, 2025

    Artificial AnalysisAgentic Index: 0.6 · Rank 60 of 64, among scored models

    Models that score at least as high for less in this scenario: 6

    $0.52 Cost of the job $0.15 · mixed / 1M tokens
  45. Llama 4 Maverick

    Meta · On OpenRouter since April 5, 2025

    Artificial AnalysisAgentic Index: 0.6 · Rank 60 of 64, among scored models

    Models that score at least as high for less in this scenario: 22

    $2.44 Cost of the job $0.5125 · mixed / 1M tokens via Parasail
  46. Llama 4 Scout

    Meta · On OpenRouter since April 5, 2025

    Artificial AnalysisAgentic Index: 0.5 · Rank 62 of 64, among scored models

    Models that score at least as high for less in this scenario: 11

    $1.05 Cost of the job $0.15 · mixed / 1M tokens via DeepInfra
  47. Gemma 3 12B

    Google · On OpenRouter since March 13, 2025

    Artificial AnalysisAgentic Index: 0.1 · Rank 63 of 64, among scored models

    Models that score at least as high for less in this scenario: 7

    $0.53 Cost of the job $0.075 · mixed / 1M tokens via DeepInfra
  48. Gemma 3 27B

    Google · On OpenRouter since March 12, 2025

    Artificial AnalysisAgentic Index: 0.1 · Rank 63 of 64, among scored models

    Models that score at least as high for less in this scenario: 9

    $0.80 Cost of the job $0.10 · mixed / 1M tokens via DeepInfra

Models that declare tool calling but have no Agentic Index score: 223 Open all models that declare tool calling in the catalog

Method and sources

How tasks are assigned
A task chip is shown only when a third party has measured the model and it ranks in the top 10% or the top 25% of the scored models. Tasks matched by a declared feature (context size, input and output types) are listed on the task page and the model page, without a chip. We do not score models ourselves. The share is rounded up to whole models, so a rank slightly past the exact share still counts. Models with equal scores share the better rank.
What "no cheaper model scores as high" means
For a task we take one third-party score and the cost of a standard job (our assumption, printed beside every cost). A model is marked when no other model scores at least as high for less. It is a computed fact, not a recommendation.
Artificial Analysis / Design Arena
Intelligence, Coding and Agentic indices: Artificial Analysis (artificialanalysis.ai). Design categories: Design Arena (designarena.ai). Both are relayed by the OpenRouter public model list and were read on November 21, 2026. We compute ranks, tiers and costs from these numbers; the sources have not reviewed them. Scores belong to the model, not to the route it is served on. The sources do not publish how many comparisons or runs stand behind each number; new models can still move.

Artificial Analysis · Design Arena · Relayed by OpenRouter. opens in a new tab

Fixed numbers we chose
Top 10% and Top 25% bands, at least 30 scored models for a band label, long documents from 400k tokens of context, budget up to $0.30 mixed price per 1M tokens, "ending soon" within 30 days.