Find the right AI model

Prices, context and performance of 360 language models from 52 providers, plus 178 models for images, video and audio. With a Fit Score for what you need them for.

Models
360
Providers
52
Updated
10/14 15:00 UTC · stored
Next sync
T−00:59:51

Model finder

Pick your use case and budget. Every model we can score is ranked by our Fit Score.

Use case

Writing, planning, brainstorming

Priority

Zero retention: the provider does not store prompts or responses.

Costs estimated for 1M input and 250K output tokens per month, API base rates.

Recommendations

196 matching models

#ModelFitCost / monthContextActions
1
ProprietaryAnthropic
85.5$4.501M
2
OpenXiaomi
85.3$0.6531.1M
3
ProprietaryAnthropic
84$0.225*1M
4
ProprietaryMeta
82.5$2.3131M
5
OpenZ.ai
82.4$0.2611M

The Fit Score factors in price and budget. For pure performance without price, see the leaderboard. Click a model name to see how its score is built. Method · Calculated from stored data.

Switch check

Is there a cheaper alternative to your current model with a similar performance index?

Which model for which task?

Pick a task: text, code, websites, images, video or speech. You see which models lead there right now, ranked by performance only. Prices are shown alongside but never change the rank.

Show

Text & reasoning

Building & design

Image, video & audio

Top 10: Overall

Index points, not a percentage. A gap of one or two points is hardly noticeable in everyday use.

#ModelPointsGapInput / output per 1MYour cost / monthActions
1
ProprietaryAnthropic
57.6Leader$4.00 / $20.00$9.00
2
ProprietaryAnthropic
56−1.6$2.00 / $10.00$4.50
3
ProprietaryAnthropic
53.4−4.2$10.00 / $50.00$22.50
4
ProprietaryOpenAI
52.7−4.9$10.00 / $50.00*$22.50
5
ProprietaryGoogle
52.6−5$2.00 / $10.00$4.50
6
ProprietaryOpenAI
51.8−5.8$2.00 / $10.00*$4.50
7
ProprietaryAnthropic
50.8−6.8$5.00 / $25.00$11.25
8
ProprietaryAnthropic
49.6−8$10.00 / $50.00$22.50
9
ProprietaryMeta
48.1−9.5$1.25 / $4.25$2.313
10
ProprietaryOpenAI
47.6−10$2.00 / $10.00*$4.50

One variant per model, free variants merged. Monthly cost for your token volume from the model finder. Data sources in the Method section.

What do you get for your money?

AI models read, write and bill in tokens, small pieces of text. Here is how much that is, and how far your budget goes.

1M tokens ≈ 750,000 words ≈ 3,000 pages ≈ 10 novels

Claude Sonnet 5.5: $2.00 per 1M input tokens, $10.00 per 1M output tokens. Preset to the top recommendation from the finder.

TaskCost per taskFor $10.00

Chat replies

A question with a short history, a detailed answer · ≈ 800 tokens

0.6¢≈ 1,800×

Email drafts

Bullet points in, a finished email out · ≈ 650 tokens

0.4¢≈ 2,200×

Document summaries

Read 10 pages, write a one-page summary · ≈ 3,800 tokens

$0.012≈ 860×

Code reviews

Review and comment on 500 lines of code · ≈ 6,500 tokens

$0.021≈ 480×

Book analyses

Analyse a 300-page book in one go · ≈ 102,000 tokens

$0.22≈ 45×

Rule of thumb for English text: 1 token ≈ 0.8 words, 1 page = 250 words, 1 novel = 300 pages, 1 line of code ≈ 10 tokens. Actual numbers vary by model, language and content; costs are base rates without caching, tiered pricing or taxes.

The leaders fly in formation

The strongest model from each of 5 providers is only 11.2 points apart. Your task decides which one fits, not the ranking.

  1. Claude Opus 5.5Anthropic57.6
  2. GPT-6 AstraOpenAI52.7
  3. Gemini 4 ArgonGoogle52.6
  4. Muse Spark 1.3Meta48.1
  5. Grok 4.7xAI46.4

Catalog

The full AgentKernel catalog, strongest models first. The Fit column also weighs price and budget for your “Everyday” profile.

/

Updating catalog … · Updated Oct 14, 2026, 15:00 UTC

SelectModelCapabilities
ProprietaryAnthropic
Reasoning
Vision
Tools
57.681.2n/an/a1M$4.00$20.00
ProprietaryAnthropic
Reasoning
Vision
Tools
5685.5n/an/a1M$2.00$10.00
ProprietaryAnthropic
Reasoning
Vision
Tools
53.4–81.657.91M$10.00$50.00
ProprietaryOpenAI
Reasoning
Vision
Tools
52.7–76.9511.1M$10.00*$50.00*
ProprietaryGoogle
Reasoning
Vision
Tools
52.682.4n/an/a1M$2.00$10.00
ProprietaryOpenAI
Reasoning
Vision
Tools
51.881.6n/an/a1.1M$2.00*$10.00*
ProprietaryAnthropic
Reasoning
Vision
Tools
50.8–7856.51M$5.00$25.00
ProprietaryAnthropic
Reasoning
Vision
Tools
49.6–76.550.71M$10.00$50.00
ProprietaryMeta
Reasoning
Vision
Tools
48.182.575.855.51M$1.25$4.25
ProprietaryOpenAI
Reasoning
Vision
Tools
47.677.8n/an/a1.1M$2.00*$10.00*
ProprietaryOpenAI
Reasoning
Vision
Tools
4777.277.450.21.1M$2.00*$10.00*
ProprietaryxAI
Reasoning
Vision
Tools
46.478.5n/an/a500K$2.00*$6.00*
OpenXiaomi
Reasoning
Vision
Tools
46.385.3n/an/a1.1M$0.435$0.87
ProprietaryQwen
Reasoning
Vision
Tools
45.477.576.2561M$2.00$6.00
OpenZ.ai
Reasoning
Tools
44.882.174.853.11M$0.079$4.80

1–15 of 360 models

Prices in USD per 1M tokens, base rate. * Tiered pricing possible. “open”: data missing, “–”: does not meet your requirements. Neither is a bad rating.

Same orbit, 10× price range

10 models are within ten points of the top of the performance index. At 1M input and 250,000 output tokens they cost between $2.313 and $22.50 per month.

News

Releases, research and reporting from provider blogs and independent trade media, original headlines linking to the publisher.

Fetching sources …

Sources: OpenAI, Google DeepMind, Google Research, Microsoft Research, Apple Machine Learning, Hugging Face, NVIDIA, AWS Machine Learning, MIT Technology Review, Ars Technica, TechCrunch, The Verge, Wired, Simon Willison, and The Decoder. Topics are derived from keywords in the headline. Announcements by providers are not independent tests.

Range: one million tokens

115 models from 25 providers accept at least 1M tokens in a single request. That is about 3,000 pages of text or 10 novels at once, enough for whole codebases or case files.

Frequently asked questions

Tokens, costs and context in plain words. The figures come from the current AgentKernel catalog.

What is a token?

A token is a small piece of text, usually part of a word. AI models read, write and bill in tokens. As a rule of thumb, one token is about 0.75 English words: 1,000 tokens are roughly 750 words or three pages.

What can I do with one million tokens?

One million tokens are about 750,000 words, roughly 3,000 pages or 10 novels. That covers, for example, about 1,300 detailed chat replies, 1,500 email drafts or around 260 summaries of ten-page documents.

How much does an AI model cost per month?

At 1M input and 250,000 output tokens per month, the 211 rated paid models in the AgentKernel catalog cost between $0.037 and $300.00, with a median of $0.90 (API base rates, as of Oct 14, 2026). Chat subscriptions are billed differently; free variants usually come with usage limits.

Which AI model is the best?

It depends on the task and the budget. The strongest model from each of 5 providers is only 11.2 points apart on the AgentKernel performance index. The model finder weighs performance, cost, context and features for your use case instead of crowning a winner.

Which AI model is best for coding?

118 models in the AgentKernel catalog have a coding index. The strongest coding model from each of 5 providers is 5.4 points apart, so price and context often decide more than rank. In the model finder, the Coding profile averages the general performance index with the coding index.

Which AI model is best for creating images?

The AgentKernel catalog lists 57 image models. In head-to-head user votes, GPT Image 2.5 Sunburst (OpenAI), GPT Image 2.5 Flare (OpenAI), GPT Image 2 (medium) (OpenAI) currently lead. Which one fits depends on style, text in the image and price; the leaderboard also ranks image editing and graphics & logos separately.

Which AI models can create videos?

30 video models are in the catalog, 17 of them with generated sound, clips up to 30 seconds and prices from $0.02 per second of video. In head-to-head user votes for video from text, Gemini Omni 1.1 Flash (Google), Gemini Omni Flash (Google), FLUX 3 Video (Black Forest Labs) currently lead. The leaderboard ranks video from text, from an image and video editing separately and shows which of them you can book via API.

Which AI voice (text to speech) sounds best?

In blind head-to-head votes, Inworld TTS MAX (Inworld), CastleFlow v1.0 (Async), Aurora (–) currently sound the most natural. The leaderboard shows the full ranking with vote counts and uncertainty, plus which models have open weights.

Which AI models can I use without my data being stored, or in the EU?

260 language models in the catalog are offered by at least one provider with zero data retention, meaning prompts and responses are not stored; 57 of them also in an EU region. 150 models have open weights and can be self-hosted. The model finder can filter for both; this is provider information, not legal advice.

What is a context window?

The context window is the largest amount of text a model can handle in a single request, measured in tokens: your question, attached documents, the conversation history and the answer together. 115 models in the catalog handle at least 1M tokens, roughly 3,000 pages at once.

What is the difference between input and output tokens?

Input tokens are the text you send to the model, output tokens the text it writes. Output is usually much more expensive: in the AgentKernel catalog the median output price is 4 times the input price. Long documents with short answers are therefore often cheaper than you might expect.

How current is the data?

The AgentKernel catalog of 360 models from 52 providers is synced every hour and the news every 15 minutes. This page reflects data as of Oct 14, 2026. Prices are the listed base rates in US dollars per million tokens.

How is the AgentKernel Fit Score calculated?

The Fit Score combines four components from 0 to 100: overall performance, budget headroom, context headroom and matching features. The weighting depends on your use case and priority; the full formula is in the Method section.

Analysis

Editorial summaries by AgentKernel as of 2 October 2026. Sources link to the original publications.

Recent developments

OpenAI presents Sol for demanding coding and knowledge work. Anthropic aims Sonnet at well-defined everyday tasks. The common trend: more specialisation within a model family.

Topic: Models & costs

Our take: compare quality and token prices separately. Then test on your own tasks whether the seemingly cheaper candidate also needs fewer attempts.

Long-term questions

What matters is whether a system reliably finishes longer, multi-step tasks. METR studies this with tasks whose difficulty is expressed as the time a human needs for them.

Status: Measurable progress

What to watch next

Success rates on longer tasks, error recovery and the number of human interventions needed.

A 50% time horizon means an estimated 50% chance of success. It is not a guaranteed autonomous runtime and not a measure of how automatable every job is.

METR: Task-Completion Time Horizons

Method

How the Fit Score (version 1.1) is built: four components from 0 to 100, weighted by use case and priority.

ComponentCalculationYour weight
PerformancePerformance index ÷ 60 × 100, capped at 100. Coding: averaged with coding index ÷ 85 × 100, Agents: averaged with agentic index ÷ 60 × 100, where available

Example: 45 index points give 75

55 %
Budget headroom100 ÷ (1 + monthly cost ÷ budget), free = 100

Example: half the budget gives 67

35 %
Context headroom70 when the minimum context is met, up to 100 at four times the minimum

Example: 128,000 with a 64,000 minimum gives 80

10 %
Features for your profileTool calling and reasoning 50 each; Documents profile: file and image input 50 each

Example: tools only gives 50

0 %
ProfilePerformanceBudgetContextFeaturesMinimum context
Everyday (your choice)55 %35 %10 %0 %16K
Coding60 %20 %10 %10 %64K
Agents50 %25 %10 %15 %32K
Documents50 %20 %25 %5 %128K
Weights per use case. Your row reflects the “Balanced” priority.
  • Priority: “Performance” shifts up to 30 percentage points from budget to performance, “Save money” up to 25 points the other way.
  • Open / no fit: “open” means the performance index, price or context size is missing. “No fit” means the context is too small, the cost is over budget, tool calling is missing for the Agents profile, image input is missing or free variants are hidden.
  • Cost: input tokens × input price + output tokens × output price, listed base rate, without caching, tiered pricing or taxes.
  • Switch check: a matching model that costs less and is at most five index points below your current model.
  • Open / proprietary: “Open” means the model weights are public on Hugging Face, so you can download and self-host the model. Not every such licence is open source in the strict sense; check it before commercial use. “Proprietary” means no public weights are listed and the model is only available through a provider API.
  • Privacy: “Zero retention” means at least one provider hosts the model with zero data retention, so prompts and responses are not stored. “In the EU” means at least one of those providers runs in an EU region. These are provider statements, not a legal review.
  • Specialist indices: not every model has a coding or agentic index. Without the matching one, the Coding or Agents profile uses the performance index alone; the model’s score breakdown says so.

Data sources: catalog, prices and capabilities from the OpenRouter Models API; performance, coding and agentic index based on the Artificial Analysis indices of the same name; design & build, image and speech placements from Design Arena; knowledge cutoff as stated by the provider; zero-data-retention hosting, EU regions and availability from OpenRouter’s endpoint list; open weights when a model is published on Hugging Face; video facts from the OpenRouter video API. We check and normalise this data but do not run the benchmarks ourselves. The Fit Score is guidance, not a task-specific test. Try your favourites on your own tasks.

Rankings for images, video and speech: LMArena Leaderboard Dataset (create and edit images, video from text, from image and editing; licence CC BY 4.0, reduced to the overall ranking, names normalised); CC BY 4.0; TTS Arena V2 (text to speech).