AI Agent Store/Model comparisonsBuild an AI worker ↗RELEASE WATCH · UPDATED 2026-11-03
Find new models, check which versions have independent results, and estimate how switching could change your bill.
Compare upgrade costs ↓Newest first. Each card labels its date as a release, announcement or catalog addition; a new hosting offer can use an older model.
Provider documentation discovered 2026-11-03
Matched LiveBench pending
Provider specifications · prices unverified
Explore scientific questions and draft research workflows for expert review. Access to the model files requires accepting the publisher’s conditions.
Released 2026-11-02
Matched LiveBench pending
Vercel AI Gateway · USD
Try LongCat 2.5 through an existing Gateway account, with the limits and billing of that hosted offer.
Released 2026-11-02
Matched LiveBench pending
LongCat native promotion · USD
Use text and pictures in coding and multi-step agent work. Check the limits of your chosen service before sending a large project.
Provider documentation discovered 2026-11-02
Matched LiveBench pending
Provider specifications · prices unverified
Consider Giotto’s smaller reasoning model when you need a private deployment on your own compatible hardware.
Provider documentation discovered 2026-11-02
Matched LiveBench pending
Giotto website USD display
Use Giotto’s hosted reasoning model for document and coding workflows, or discuss a private deployment with its team.
Provider documentation discovered 2026-11-02
Matched LiveBench pending
OpenAI promotional API rates · USD
A legal-work configuration now listed by OpenAI. Confirm access and try it on documents you can check before planning a rollout.
Released 2026-11-01
Matched LiveBench pending
Native Drex API · USD
Check financial records against rules and answers you define—for example, whether an invoice exceeds its approved purchase order.
Released 2026-10-30
Matched LiveBench pending
Provider specifications · prices unverified
Run a downloadable text-and-image model for document questions, coding and tool-assisted work. Its unusual design needs compatible serving software.
Released 2026-10-30
Matched LiveBench pending
Temporary Gateway promotion · USD
Use Ling 1019 through Gateway, but confirm billing before leaving an application running: its announced free period ends today.
Released 2026-10-30
Matched LiveBench pending
Provider specifications · prices unverified
Try a downloadable coding and reasoning model on your own hardware. This preview needs its supplied serving code and substantial memory.
Released 2026-10-30
Matched LiveBench pending
Restricted API, short context · USD
Help approved life-sciences teams analyze evidence, plan experiments and work with scientific tools.
Announced 2026-10-30
Matched LiveBench pending
Provider specifications · prices unverified
A planned model for choosing from defined answers, such as routing a ticket or checking whether a condition is met.
Released 2026-10-29
Matched LiveBench pending
Provider specifications · prices unverified
Run ticket routing, classification or yes/no checks on your own hardware, with probabilities for the answers you define.
Released 2026-10-29
Matched LiveBench pending
Provider specifications · prices unverified
Run ticket routing, classification or yes/no checks on your own hardware, with probabilities for the answers you define.
Released 2026-10-29
Matched LiveBench pending
AI Gateway · USD
Route support tickets or rate text records against choices you define. This Gateway offer lists text input; use the native API if you need image input.
Released 2026-10-29
Matched LiveBench pending
Native Baseten API · USD
Route support tickets, check a screenshot or rate a record against choices you define. Joey-1 is the small, low-price member of Base Labs’ decision family.
Announced 2026-10-29
Matched LiveBench pending
Provider specifications · prices unverified
Extract text, tables and page locations from difficult documents, so an assistant can point readers to the source passage.
Announced 2026-10-29
Matched LiveBench pending
Provider specifications · prices unverified
A smaller decision model is planned. Wait for the exact download and supported limits before choosing it.
Announced 2026-10-29
Matched LiveBench pending
Provider specifications · prices unverified
Run simple ticket routing or topic classification locally with a small download. Test more demanding judgment tasks separately; small size does not imply comparable reasoning quality.
Released 2026-10-28
Matched LiveBench pending
Native Drex API · USD
Answer up to 512 defined questions about text and images in one request. Use this version when the full set fits within one shared token budget.
Released 2026-10-28
Matched LiveBench pending
OpenRouter offer · USD
Try a coding model trained to finish software tasks with fewer generated tokens. Compare finished work, retries and the whole bill on your own repositories.
Announced 2026-10-28
Matched LiveBench pending
Announced introductory rates · USD
Watch Carbon for demanding software-engineering and cybersecurity work. Access starts with selected defenders and trusted partners; most developers cannot switch to it yet.
Released 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Use this pretrained foundation for further training and model research. It is not the ready-to-chat Flash 1019 model.
Released 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Classify a voice note or ask spoken questions about images and records, without generating a written reply.
Released 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Route a support ticket, read a value from a document or locate objects in an image using a model you run yourself.
Released 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Ask a spoken question about a photo, or let a local assistant see and hear a video together.
Released 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Build a local voice assistant that transcribes, translates, answers aloud or turns spoken requests into tool calls.
Released 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Read documents, charts and app screens, or ask questions about a video on your own device.
Released 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Run a small text assistant on your own hardware, connect it to tools or adapt it to your business vocabulary.
Released 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Download this revised Ling model for coding and tool-based workflows. Test the exact revision before replacing the earlier Flash model.
Provider documentation discovered 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Explore plans for a changing physical scene by combining written reasoning with generated video predictions. This is a research setup for scene understanding and visual planning.
Provider documentation discovered 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Find, count or outline objects in a still image, locate text, or separate a subject from its background. Choose the 2B or 4B download and use Tencent’s inference package.
Provider documentation discovered 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Route support requests, prioritize cases or check an image against a rule on your own machine. Supply the possible answers; d3 scores them instead of writing a reply.
Provider documentation discovered 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Read complex document tables, charts and formatting through LlamaParse. For example, recover a financial table while keeping each value attached to its row and column.
Provider documentation discovered 2026-10-27
Matched LiveBench pending
Provider specifications · prices unverified
Run a downloadable model for coding, document work and tool-assisted tasks. It is intended both for applications and for training smaller models.
Provider documentation discovered 2026-10-27
Matched LiveBench pending
Vercel AI Gateway · USD
Run Terra through Gateway’s faster-processing offer. Check whether reduced waiting is worth the higher bill on your own workload.
Released 2026-10-26
Matched LiveBench pending
OpenRouter offer · USD
Try Terra for reading screenshots, comparing product images, working in apps or revising text. Its standard token prices are half Sol’s; compare the results on your own tasks before switching.
Provider documentation discovered 2026-10-26
Matched LiveBench pending
Provider specifications · prices unverified
Choose a smaller download format for running Qwen3.8 Flash Next locally. S, M and L offer different memory requirements; compare them on your own mix of writing and coding.
Provider documentation discovered 2026-10-26
Matched LiveBench pending
Provider specifications · prices unverified
Build an Android assistant that reads screenshots and takes actions in apps. Your application connects the model to the phone and decides which actions it may execute.
Provider documentation discovered 2026-10-26
Matched LiveBench pending
Provider specifications · prices unverified
Build an Android assistant that reads screenshots and takes actions in apps. Your application connects the model to the phone and decides which actions it may execute.
Catalog added 2026-10-26 UTC
Matched LiveBench pending
OpenRouter offer · USD
OpenRouter serves the Terra model with extra reasoning in pro mode. It can use substantially more tokens; this is a hosting option, not a separately verified base model.
Provider documentation discovered 2026-10-25
Matched LiveBench pending
Liquid native API · USD
Classify a record or image, choose from allowed answers, or score a case against your rules. This is now Liquid’s default hosted decision model.
Released 2026-10-23
Matched LiveBench pending
AI Gateway · USD
Consider the higher-priced service when response time matters. Test the time saved on your real requests before paying more.
Released 2026-10-23
Matched LiveBench pending
Provider specifications · prices unverified
Run a small text assistant locally, including tool calls. A voice assistant still needs separate speech recognition and speech output.
Released 2026-10-23
Matched LiveBench pending
OpenRouter offer · USD
Try the K3 update for coding and long-document work. Compare it on your own tasks before replacing K3; independent results are still pending.
Provider documentation discovered 2026-10-23
Matched LiveBench pending
Provider specifications · prices unverified
Let Fireworks choose a model for each turn of coding work. If you need one fixed model, check whether your plan supports pinning it.
Provider documentation discovered 2026-10-23
Matched LiveBench pending
Provider specifications · prices unverified
Make local choices for games, web-app testing, multilingual tasks and documents. This revision changes the underlying model as well as adding specialists, so retest existing workflows before upgrading.
Provider documentation discovered 2026-10-23
Matched LiveBench pending
Provider specifications · prices unverified
Classify text locally using fixed choices or a scoring scale you define. This v3 package splits long question lists while keeping the full shared information in each part.
Provider documentation discovered 2026-10-23
Matched LiveBench pending
Z.ai native API · USD
A specialized option for legal research and document workflows in English and Chinese. Check its answers against the underlying documents before relying on them.
Provider documentation discovered 2026-10-23
Matched LiveBench pending
Provider specifications · prices unverified
Run a coding model on your own hardware to inspect a codebase, make changes and write tests.
Released 2026-10-21
Matched LiveBench pending
Provider specifications · prices unverified
Try local coding on a 24GB or 32GB graphics card. These compressed, coding-focused versions of Qwen3.8 Flash Next trade broader knowledge for a smaller memory requirement.
Released 2026-10-21
Matched LiveBench pending
Provider specifications · prices unverified
Run fixed-choice classification and scoring on your own hardware, with text or images.
Released 2026-10-21
Matched LiveBench pending
AI Gateway · USD
Choose from defined answers, route records or score a condition. Compare this model on your own examples before automating decisions.
Released 2026-10-21
Matched LiveBench pending
Native Baseten API · USD
Choose from defined answers, route records or score a condition. Compare this model on your own examples before automating decisions.
Released 2026-10-21
Matched LiveBench pending
Provider specifications · prices unverified
Run fixed-choice classification and scoring on your own hardware, with text or images.
Released 2026-10-21
Matched LiveBench pending
AI Gateway · USD
Choose from defined answers, route records or score a condition. Compare this model on your own examples before automating decisions.
Released 2026-10-21
Matched LiveBench pending
Native Baseten API · USD
Choose from defined answers, route records or score a condition. Compare this model on your own examples before automating decisions.
Provider documentation discovered 2026-10-21
Matched LiveBench pending
Cloudflare Workers AI · USD
Use the October 17 DeepSeek model through Cloudflare Workers AI for text, images and tool-assisted work.
Provider documentation discovered 2026-10-21
Matched LiveBench pending
Vercel AI Gateway · USD
Summarize a long document or transcript with instructions for language, length and focus.
Provider documentation discovered 2026-10-21
Matched LiveBench pending
Vercel AI Gateway · USD
Extract fields from a document using the structure you define, such as invoice amounts or contract parties.
Provider documentation discovered 2026-10-21
Matched LiveBench pending
Provider specifications · prices unverified
Turn a request such as “find flights tomorrow” into a category and fields your application can use. Supply your categories, field descriptions and text or conversation.
Provider documentation discovered 2026-10-21
Matched LiveBench pending
Provider specifications · prices unverified
Sort messages or select a next action locally using answer choices you define. This is Labqoat’s model; similarly named Qev downloads from other authors are separate.
Provider documentation discovered 2026-10-21
Matched LiveBench pending
Provider specifications · prices unverified
Run a large reasoning model yourself for coding and tool-assisted work. Choose the weight format to fit your hardware before planning a deployment.
Provider documentation discovered 2026-10-21
Matched LiveBench pending
Provider specifications · prices unverified
Run a compact model locally for coding, text tasks and tool-assisted workflows. It builds on Spark X2.5 with additional training.
Catalog added 2026-10-21 UTC
Matched LiveBench pending
OpenRouter offer · USD
Try Flint for coding and tool-assisted text tasks through a currently free OpenRouter offer. Check availability before building around a free service.
Provider documentation discovered 2026-10-20
Matched LiveBench pending
Cloudflare Workers AI · USD
Turn an invoice or screenshot into useful fields, such as an amount, due date or estimated payment time. Missing values and uncertain predictions need different handling.
Provider documentation discovered 2026-10-20
Matched LiveBench pending
Native launch discount · USD
Route a support request, choose a category or score urgency. Send shared information and named questions; the service returns probabilities and predefined answers.
Released 2026-10-19
Matched LiveBench pending
OpenRouter · USD
Use the stable model through a paid OpenRouter offer after the free route disappeared.
Released 2026-10-19
Matched LiveBench pending
MiniMax direct API · USD
Use text, screenshots and video in coding or multi-step tasks. This is the stable October release, separate from the earlier subscription preview.
Released 2026-10-19
Matched LiveBench pending
OpenRouter offer · USD
A text model designed for responsive chat and tool-driven tasks. Compare it on your own examples before choosing it for a speed-sensitive workflow.
Released 2026-10-19
Matched LiveBench pending
OpenRouter offer · USD
This free OpenRouter offer is no longer listed. The stable model is available through a separate paid offer.
Provider documentation discovered 2026-10-19
Matched LiveBench pending
AI Gateway / Baseten offer · USD
Run the October 17 Flash model through Baseten on AI Gateway. This option supports long inputs and cached prompts. Test response times on your own workload before paying more for the Fast offer.
Released 2026-10-17
Matched LiveBench pending
Provider specifications · prices unverified
Turn document images into readable text, tables and chart data while retaining reading order and locations on the page.
Released 2026-10-17
Matched LiveBench pending
OpenRouter offer · USD
The October 17 update is available for coding, visual questions and work with tools. DeepSeek reports faster, shorter answers, but independent results for this exact revision are still pending.
Catalog added 2026-10-17 UTC
Matched LiveBench pending
OpenRouter offer · USD
Ask questions about charts, images and video. This hosting option has different context and output limits from Gateway; choose the service that fits your workload.
Released 2026-10-16
Matched LiveBench pending
AI Gateway offer · USD
Analyze charts, follow objects in video or count items in images. Glimpse is a research preview for visual questions; its maker’s comparisons use tools and settings that differ across models, so they are not a like-for-like quality ranking here.
Released 2026-10-16
Matched LiveBench pending
Direct Decisions API · USD
A lower-priced option for large volumes of clear decisions, such as labeling messages or routing tickets. Test it against examples you have already labeled before choosing it on price.
Released 2026-10-16
Matched LiveBench pending
Direct Decisions API · USD
A decision model for checking claims, grading records against a rubric or routing difficult requests. It returns probabilities and predefined choices, rather than a written answer.
Released 2026-10-16
Matched LiveBench pending
OpenRouter offer · USD
A paid hosted offer for Beam’s coding and tool-based workflows. Compare its ongoing cost with Gateway’s temporary free access; neither option has a matched independent quality result here.
Provider documentation discovered 2026-10-16
Matched LiveBench pending
Cloudflare Workers AI · USD
Generate replies and call tools inside a voice assistant. This Cloudflare offer provides the text model; speech recognition and spoken output need their own components.
Provider documentation discovered 2026-10-16
Matched LiveBench pending
Perplexity Agent API · USD
Use this text-only Nemotron reasoning offer through Perplexity’s Agent API. It replaces the older Ultra offer in that service’s model list.
Provider documentation discovered 2026-10-16
Matched LiveBench pending
AI Gateway offer · USD
Choose a fixed Jev version when you need repeatable ticket routing, labels or scores. Retest your confidence threshold before changing versions.
Catalog added 2026-10-16 UTC
Matched LiveBench pending
OpenRouter offer · USD
Use a compact text-and-image model through OpenRouter, or download it for a compatible local setup. Its compressed files reduce storage needs; test quality and total memory on your own task.
Catalog added 2026-10-16 UTC
Matched LiveBench pending
OpenRouter offer · USD
An open-weight text model for coding, document questions and tool-based workflows. This newly listed free service is one way to try it; running your own copy still needs hardware. Independent comparison results are pending.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Describe an action in a selected part of a robot or human demonstration video. For example, turn a clip into an instruction that can label a training example.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Find where one action ends and another begins in a robot or human demonstration video. Use the results to prepare a robotics research dataset.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Build an Arabic and English text assistant, including regional Arabic and Arabizi. Test it on the dialects, documents and tool workflows your users actually need.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Build an Arabic and English text assistant, including regional Arabic and Arabizi. Test it on the dialects, documents and tool workflows your users actually need.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Build an Arabic and English text assistant, including regional Arabic and Arabizi. Test it on the dialects, documents and tool workflows your users actually need.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Build an Arabic and English text assistant, including regional Arabic and Arabizi. Test it on the dialects, documents and tool workflows your users actually need.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Build an Arabic and English text assistant, including regional Arabic and Arabizi. Test it on the dialects, documents and tool workflows your users actually need.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Build an Arabic and English text assistant, including regional Arabic and Arabizi. Test it on the dialects, documents and tool workflows your users actually need.
Released 2026-10-15
Matched LiveBench pending
Unbiased native API · USD
Send a request to a service that combines several models into one answer. This is the stable October release; the older OpenRouter preview page is a separate offer.
Released 2026-10-15
Matched LiveBench pending
Direct Liquid API · USD
Check a photo against a damage claim, inspect a component or route a difficult support case. Supply the possible answers; the model returns probabilities rather than writing a reply.
Released 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Run text and visual tasks on your own hardware with NVIDIA’s compressed 74B model. The 24GB package reduces the weight footprint; working memory and a compatible serving setup are still needed.
Released 2026-10-15
Matched LiveBench pending
OpenRouter offer · USD
Read documents, screenshots and video, then produce text or call tools. This new NVIDIA model has no matched independent score in this comparison yet.
Released 2026-10-15
Matched LiveBench pending
OpenRouter offer · USD
Use Meta’s downloadable model for coding and workflows that read text, images, audio or video. A hosted API avoids the large hardware investment needed to run it yourself.
Provider documentation discovered 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
Run the v1.1 decision model on your own GPU to assign labels and probabilities to text or images. This download has different limits from the hosted Gateway offer.
Provider documentation discovered 2026-10-15
Matched LiveBench pending
Provider specifications · prices unverified
This older Gateway offer is no longer in the current catalog. For a new project, start with Decider v1.2 and test it on decisions you have already labeled.
Catalog added 2026-10-15 UTC
Matched LiveBench pending
OpenRouter offer · USD
A hosted option for the stable October service, which combines several models into one answer. Its token prices match the published native quote; access and data handling still depend on the service you choose.
Released 2026-10-14
Matched LiveBench pending
Sqwish base model · USD
Route a request, check a record or choose an action from answers you define. Use your own labeled examples to choose the smallest model that works well enough.
Released 2026-10-14
Matched LiveBench pending
Sqwish base model · USD
Route a request, check a record or choose an action from answers you define. Use your own labeled examples to choose the smallest model that works well enough.
Released 2026-10-14
Matched LiveBench pending
Sqwish base model · USD
Route a request, check a record or choose an action from answers you define. Use your own labeled examples to choose the smallest model that works well enough.
Released 2026-10-14
Matched LiveBench pending
Provider specifications · prices unverified
Run chat, classification and speech transcription on one server. Separate built-in components handle general decisions and harmful-content checks, so you can reuse the same deployment for several steps of a workflow.
Released 2026-10-14
Matched LiveBench pending
Provider specifications · prices unverified
Run chat, classification and speech transcription on one server. Separate built-in components handle general decisions and harmful-content checks, so you can reuse the same deployment for several steps of a workflow.
Released 2026-10-14
Matched LiveBench pending
AI Gateway offer · USD
Choose a fixed Jev version when you need repeatable ticket routing, labels or scores. Retest your confidence threshold before changing versions.
Released 2026-10-14
Matched LiveBench pending
Provider specifications · prices unverified
Choose labels or yes/no answers for records in English, Hindi, Tamil and Kannada. This downloadable model returns probabilities and can serve a ticket-routing workflow on your own hardware.
Announced 2026-10-14
Matched LiveBench pending
Provider specifications · prices unverified
An announced small text decision model intended for local use. A usable download and public API ID are not yet verified.
Provider documentation discovered 2026-10-14
Matched LiveBench pending
AI Gateway offer · USD
Assign text to your own labels, such as sales, support or billing, and receive a probability for each label.
Provider documentation discovered 2026-10-14
Matched LiveBench pending
Provider specifications · prices unverified
Google offers a specialized Argon version to approved cyber defenders for finding vulnerabilities and preparing fixes.
Released 2026-10-13
Matched LiveBench pending
Native Drex API · USD
Check text and images against up to 512 questions—for example, identify a damaged parcel and choose how to handle the claim.
Released 2026-10-13
Matched LiveBench pending
OpenRouter offer · USD
A text model for assistants that reason and use tools. You can use a hosted service or download the files to run it on your own hardware.
Released 2026-10-13
Matched LiveBench pending
OpenRouter offer · USD
Use Argon for demanding coding, research and tasks with several steps. Public API access began October 13. Choose low, medium or high thinking; medium is the default.
Provider documentation discovered 2026-10-13
Matched LiveBench pending
Provider specifications · prices unverified
Run typed decisions on your own hardware. This 35B model uses about 3B parameters per token; it returns option probabilities rather than prose.
Catalog added 2026-10-13 UTC
Matched LiveBench pending
OpenRouter offer · USD
Turn a long document into a shorter summary, with a focus or format you specify. Check that important details survive before using it in another workflow.
Catalog added 2026-10-13 UTC
Matched LiveBench pending
OpenRouter offer · USD
Pull fields such as invoice numbers, dates and customer names from documents. Describe the fields you want before sending the text.
Released 2026-10-12
Matched LiveBench pending
Provider specifications · prices unverified
Meta’s higher-capability Wonder variant for work in Muse Code. It requires a Power Usage or Maximum Usage subscription.
Released 2026-10-12
Matched LiveBench
OpenRouter offer · USD
For complex coding and tasks that need several steps. One independent LiveBench result tests max thinking; it does not tell us whether lower settings are faster or better value.
Provider documentation discovered 2026-10-11
Matched LiveBench pending
Workers AI documented rates · USD
Choose from your own answer options using text, images, audio or video. Clef Omni returns a confidence score for each answer instead of writing a paragraph.
Provider documentation discovered 2026-10-11
Matched LiveBench pending
Provider specifications · prices unverified
Run this decision model on your own hardware. Give it background information and questions with fixed answer options; it returns confidence scores instead of prose. It is based on Efficient-DLM-8B.
Provider documentation discovered 2026-10-11
Matched LiveBench pending
Provider specifications · prices unverified
Run the downloadable 9B Drex model on your own hardware. It uses the same fixed-answer format as the online service, but has separate hardware, license and setup requirements.
Provider documentation discovered 2026-10-10
Matched LiveBench pending
Provider specifications · prices unverified
This second Darwin release adds training for choosing actions and workflow steps. Test it separately from version 1, including whether its confidence scores match how often it is right.
Provider documentation discovered 2026-10-10
Matched LiveBench pending
Provider specifications · prices unverified
A downloadable text model that scores your fixed answer choices in one processing pass per question.
Released 2026-10-09
Matched LiveBench pending
Microsoft base rate / matching Gateway quote · USD
Classify a ticket, choose a workflow step or grade an answer against your rubric. Microsoft’s new text-only decision model returns probabilities over the options you supply, rather than writing a response.
Released 2026-10-08
Matched LiveBench pending
Provider specifications · prices unverified
A coding model you can run on your own hardware. Use it to investigate a failing test, draft a fix and check the result while keeping repository access under your control.
Announced 2026-10-08
Matched LiveBench pending
Provider specifications · prices unverified
Read page images into text, tables and labeled layout regions. This is the larger checkpoint; test it against the smaller versions on your documents.
Announced 2026-10-08
Matched LiveBench pending
Provider specifications · prices unverified
Read page images into text, tables and labeled layout regions. This variant keeps the earlier LightOnOCR-2 architecture.
Announced 2026-10-08
Matched LiveBench pending
Provider specifications · prices unverified
Read page images into text, tables and labeled layout regions. This is the smallest checkpoint in the family.
Catalog added 2026-10-08 UTC
Matched LiveBench pending
OpenRouter offer · USD
A new routed offer for StepFun’s coding and knowledge-work model. Native documentation supports text, image and video input, tools, 1M context and 64K output. The Gateway output allowance differs; check your exact endpoint. No matched independent score is assigned.
Released 2026-10-07
Matched LiveBench pending
Provider specifications · prices unverified
An experimental local decision checkpoint for text with images or audio. It can route spoken requests without generating a transcript or free-form answer.
Released 2026-10-07
Matched LiveBench pending
Provider specifications · prices unverified
A local decision checkpoint for text and images. It returns fixed answers in one pass rather than writing a response.
Released 2026-10-07
Matched LiveBench
OpenRouter offer · USD
Five native effort levels with medium default; thinking is on by default. The 1M context and 128K output limits are documented. Prompts above 100K use five times the base token rates. Its newer tokenizer can count the same text differently from Haiku 4.5; compare actual billed usage before switching.
Compare tested effort levels ↗Announced 2026-10-07
Matched LiveBench pending
Provider specifications · prices unverified
A research model that reads and writes bytes instead of ordinary subword tokens. Ai2’s October 7 announcement highlights this new checkpoint; repository creation and modification dates are in August, so announcement and artifact dates stay separate.
Announced 2026-10-07
Matched LiveBench pending
Provider specifications · prices unverified
A research model that reads and writes bytes instead of ordinary subword tokens. Ai2’s October 7 announcement highlights this new checkpoint; repository creation and modification dates are in August, so announcement and artifact dates stay separate.
Released 2026-10-06
Matched LiveBench pending
Native Drex API · USD
Choose from allowed answers without generating a paragraph: route a ticket, classify a document or return a probability for each option.
Released 2026-10-06
Matched LiveBench pending
Direct Decisions API · USD
Answer yes/no questions, choose named options or score records against your rules. Shared documents can be reused across requests, and each question can include its own images.
Released 2026-10-06
Matched LiveBench
OpenRouter offer · USD
Preview API access is available; open weights remain planned for later October. The native model page no longer shows its launch discount and lists $1.36 input, $0.14 cached input and $4.18 output per million. This routed offer has its own current prices.
Announced 2026-10-06
Matched LiveBench pending
Provider specifications · prices unverified
A text model adapted for Emirati Arabic, including local vocabulary, idioms and cultural context. It can be useful when a general Arabic assistant answers in the wrong dialect.
Released 2026-10-05
Matched LiveBench pending
Earlier native offer · current rate unverified
The earlier Liquid decision model remains listed by hosting services. For a new direct Liquid integration, check the newer default d1.1.
Released 2026-10-05
Matched LiveBench pending
Provider specifications · prices unverified
A text model for coding and multi-step work with tools. You can now download it under Apache 2.0 or use a hosted service. Choose hosting for a simpler start; running the full model yourself requires substantial hardware.
Announced 2026-10-05
Matched LiveBench pending
Provider specifications · prices unverified
A research preview for understanding and generating text, images and video. We have not independently verified general-task quality, response time or billing here.
Provider documentation discovered 2026-10-05
Matched LiveBench pending
Direct restricted API · USD
A specialized life-sciences reasoning model for approved research organizations. OpenAI lists October 5 as its billing start; this is a pricing milestone, not a new model launch. Comparable independent benchmarks are pending.
Released 2026-10-03
Matched LiveBench pending
Provider specifications · prices unverified
A German and English reasoning model with tool support and downloadable model files. Comparable independent general-task results and hosted token prices are still missing.
Provider documentation discovered 2026-10-03
Matched LiveBench pending
AI Gateway routed offer · USD
AI Gateway offers this version of V4.1 Flash for faster serving. We do not yet have comparable speed measurements, and results for the standard service are not assigned to this version.
Provider documentation discovered 2026-10-03
Matched LiveBench pending
Boundless via AI Gateway · USD
Route records, choose labels or score a condition using answers you define. Laya is a paid decision service on Gateway; the free promotion has ended.
Released 2026-10-02
Matched LiveBench pending
Provider specifications · prices unverified
A model for turning a research question and supplied literature excerpts into a cited report. Its published development studies do not compare today’s frontier models.
Provider documentation discovered 2026-10-02
Matched LiveBench pending
Direct Decisions API · USD
This older API name now runs Decider v1.2. Use it to classify records or choose among answers you supply, but recheck any rules that rely on the model’s probabilities after the version change.
Provider documentation discovered 2026-10-02
Matched LiveBench pending
AI Gateway paid offer · USD
A text model for coding, analysis and tools. The Gateway free promotion has ended; its standard offer now has published token prices.
Released 2026-10-01
Matched LiveBench pending
AI Gateway offer · USD
A model trained for React Native and React engineering, with Next.js use described by the provider. The native pinned version is October 1; Gateway availability is a separate October 8 offer. Comparable independent coding results are pending.
Released 2026-10-01
Matched LiveBench pending
Workers AI documented rates · USD
Use text and images to choose from answers you define. The model files are available under Apache 2.0, and Cloudflare offers an online decision service. We have not independently reproduced the provider’s accuracy or speed comparisons.
Released 2026-10-01
Matched LiveBench pending
Workers AI documented rates · USD
Use text and images to choose from answers you define. The model files are available under Apache 2.0, and Cloudflare offers an online decision service. We have not independently reproduced the provider’s accuracy or speed comparisons.
Released 2026-10-01
Matched LiveBench pending
OpenRouter offer · USD
The OpenRouter preview offer is no longer listed. Unbiased has released stable Pareto 26.10 through its own service.
Provider documentation discovered 2026-10-01
Matched LiveBench pending
Provider specifications · prices unverified
The earlier M Plan and MiniMax Code preview. The stable M3.1 Flash release has its own page; do not use this preview’s identity for new tests.
Catalog added 2026-10-01 UTC
Matched LiveBench pending
OpenRouter offer · USD
This reasoning and tool-use model’s free OpenRouter offer is no longer listed. Confirm another available offer before connecting an application.
Released 2026-09-29
Matched LiveBench
OpenRouter offer · USD
Five supported effort levels: low, medium (the API default), high, extra high and max. None and minimal are unsupported. Tool calling uses Responses; Chat Completions supports requests without tools.
Compare tested effort levels ↗Catalog added 2026-09-29 UTC
Matched LiveBench pending
OpenRouter offer · USD
OpenRouter describes this as GPT-6.1 Sol served in pro mode, with greater token use. It is a serving variant; no independent result is assigned from the base model.
Released 2026-09-28
Matched LiveBench pending
Native Drex API · USD
Send a document to Nace’s online Drex service and ask it to choose an answer, route a ticket or score a condition. It returns the requested decision instead of a paragraph.
Released 2026-09-28
Matched LiveBench pending
Provider specifications · prices unverified
This former Opus router is no longer available. Fireworks now documents Nexus Auto for model selection, or a specific model ID to keep every turn on one model.
Released 2026-09-28
Matched LiveBench
OpenRouter offer · USD
Adaptive thinking, high default effort and 128K maximum output. Migration changes affect forced tool use, thinking replay and older computer-use tools.
Compare tested effort levels ↗Released 2026-09-25
Matched LiveBench pending
OpenRouter offer · USD
Provider verifies text, image, audio and video input and 32K context; this catalog offer advertises 36,864 tokens. Confirm the selected endpoint limit. Specialized performance is not scored using another model’s text benchmarks.
Released 2026-09-23
Matched LiveBench pending
OpenRouter offer · USD
Research preview announced September 23 with temporary two-week serverless access. Vendor claims about shorter reasoning are not independently measured task costs; verify availability.
Released 2026-09-23
Matched LiveBench pending
OpenRouter offer · USD
A text-generation model listed in the provider catalog. We have not verified its general-task or storytelling performance in comparable independent tests.
Released 2026-09-23
Matched LiveBench pending
OpenRouter offer · USD
A lower-priced text-generation service. Test answer quality and retries before assuming it will cost less to complete your tasks.
Catalog added 2026-09-23 UTC
Matched LiveBench pending
OpenRouter offer · USD
A separately listed service for GLM 5.3. We have not verified its advertised speed with comparable measurements or assigned it the standard model’s benchmark results.
Catalog added 2026-09-23 UTC
Matched LiveBench pending
OpenRouter offer · USD
A separately listed service for Qwen3.8 Max. Comparable benchmark results and measured speed for this specific version are not confirmed.
Catalog added 2026-09-21 UTC
Matched LiveBench pending
OpenRouter offer · USD
Switchyard chooses from a configured group of models. OpenRouter says each request uses the selected model’s price and input capacity. There is no single confirmed fixed rate or guaranteed one-million-token capacity for every request; catalog placeholders do not mean it is free.
Released 2026-09-17
Matched LiveBench pending
OpenRouter offer · USD
Pareto combines answers from several models for research, coding and multi-step work. This named September version is newly covered here; it is not an October base-model launch.
Announced 2026-09-16
Matched LiveBench pending
AI Gateway decision offer · USD
Choose a label, score or yes/no probability for a support ticket, document or other text. Jev is a decision service rather than a chatbot.
Catalog added 2026-09-04 UTC
Matched LiveBench pending
OpenRouter offer · USD
This health-focused text model’s OpenRouter offer is no longer listed. Current access is unverified, and its earlier domain claims do not establish clinical reliability.
Released 2026-08-11
Matched LiveBench pending
AI Gateway free offer · USD
A newly listed free Gateway offer for NVIDIA’s August model. It supports text assistants, tools and long documents.
Catalog added 2026-08-11 UTC
Matched LiveBench pending
OpenRouter offer · USD
A small reasoning model for retrieval, extraction and tools. The publisher does not recommend it for agentic coding or knowledge-heavy tasks. This is newly discovered hosting, not a newly dated model release.
Catalog added 2026-08-11 UTC
Matched LiveBench
OpenRouter offer · USD
A smaller open model for text tasks and tools. LiveBench now includes its own measured result; its exact thinking budget is not specified in the result ID.
Released 2026-07-24
Matched LiveBench pending
Provider specifications · prices unverified
An openly documented multilingual model for text, images and experimental audio input. Consider it when you want to run or adapt a model yourself and inspect how it was trained.
Released 2026-07-24
Matched LiveBench pending
Provider specifications · prices unverified
An openly documented multilingual model for text, images and experimental audio input. Consider it when you want to run or adapt a model yourself and inspect how it was trained.
Released 2026-07-07
Matched LiveBench pending
OpenRouter offer · USD
Describe images, locate objects or answer questions about a picture. This is newly covered on our site; the model itself was released earlier.
Released 2025-09-18
Matched LiveBench pending
OpenRouter offer · USD
Describe images, locate objects or answer questions about a picture. This is newly covered on our site; the model itself was released earlier.
New: compare decision models for routing and classification ↗
Compare Standard, Fast and Ultrafast costs for GPT-6.1 Sol and Astra ↗
Check model routes that disappeared from a catalog ↗
On AI Gateway, K3.1 Pro Highspeed costs twice as much for uncached input and output as standard K3.1 Pro. Both quote the same price for cached input. For 1,000 requests using 10,000 uncached input tokens and 2,000 output tokens each, the listed token charges are $40 for standard and $80 for Highspeed. Thinking tokens count toward output; cache writes and other fees are extra.
Measure completed-task time before switching. A faster service can be useful for an interactive assistant, while a background job may gain little from it. The name Highspeed and the publisher’s speed claim do not establish how much time your requests will save.
Data correction · nvidia/switchyard. OpenRouter corrected the previous free-service FAQ and clarified that context is the serving model’s limit. The routing page preserves unknown fixed prices and exact system identity.
Data correction · mistral-large-4-high. LiveBench revised six task values in the existing Large 4 high row. Today’s category scores use the corrected row; the October 7 download preserves its original values. This is a benchmark correction, not a new model release or effort sweep.
Data correction · @cf/cloudflare/clef-flash and @cf/cloudflare/clef. Current exact Workers AI pages correct stale profile facts: Flash is24,576tokens/$0.038input, and both Clef endpoints explicitly charge no output tokens. This is a source-backed correction to our data; a provider price-change date is unverified. Immutable prior downloads retain the old observation.
SAME SERVICE · PRICES CHECKED TWICE
A service such as OpenRouter can change its price without the model maker changing its own price. These quotes show what we found on each date, per million tokens. Prompt length, time of use, the provider handling the request and other fees can affect your bill.
2026-11-02 → 2026-11-03
This hosting offer changed price. Check the service you will use; direct-provider prices may differ.
Check the current offer ↗2026-11-02 → 2026-11-03
This hosting offer changed price. Check the service you will use; direct-provider prices may differ.
Check the current offer ↗ACCESS AND BILLING
Deprecated; retirement scheduled
claude-sonnet-4-5-20250929
Provider-recommended replacement: claude-sonnet-5-5.
Claude API retirement notice. Partner-operated Bedrock and Google Cloud schedules can differ. Historical benchmarks remain historical; scores are not transferred to the recommended replacement.
Read the provider notice ↗Catalog expiration scheduled
z-ai/glm-4.5
Replacement recommendation: Not verified.
The exact OpenRouter entry now declares this expiration date. This is endpoint-specific catalog evidence; it does not establish a direct-provider shutdown or retirement of every hosted copy. No exact cutoff time or replacement is verified.
Read the endpoint catalog ↗Catalog expiration scheduled
z-ai/glm-4.7
Replacement recommendation: Not verified.
The exact OpenRouter entry now declares this expiration date. This is endpoint-specific catalog evidence; it does not establish a direct-provider shutdown or retirement of every hosted copy. No exact cutoff time or replacement is verified.
Read the endpoint catalog ↗Catalog expiration scheduled
dots-studio/dots-3-note-preview:free
Replacement recommendation: Not verified.
The exact OpenRouter entry now declares this expiration date. This is endpoint-specific catalog evidence; it does not establish a direct-provider shutdown or retirement of every hosted copy. No exact cutoff time or replacement is verified.
Read the endpoint catalog ↗Deprecated; retirement scheduled
gpt-5.3-codex
Provider-recommended replacement: gpt-6-sol.
Official OpenAI API notice gives six months of notice. Historical results remain attached to the tested model; no successor scores are transferred.
Read the provider notice ↗Deprecated; retirement scheduled
gpt-5.4-nano
Provider-recommended replacement: gpt-6-luna.
Official OpenAI API notice gives six months of notice. Historical results remain attached to the tested model; no successor scores are transferred.
Read the provider notice ↗Deprecated; retirement scheduled
gpt-5.1
Provider-recommended replacement: gpt-6-sol.
Official OpenAI API notice gives six months of notice. Historical results remain attached to the tested model; no successor scores are transferred.
Read the provider notice ↗Hosted endpoint deprecated; retirement scheduled
zai-glm-5-2
Provider-recommended replacement: zai-glm-5-3.
This notice applies to the Mistral-hosted endpoint. It does not retire every Z.ai GLM5.2 deployment. The provider states the replacement uses the same price; no benchmark score is transferred.
Read the provider notice ↗Managed agent · earliest shutdown date reached
antigravity-preview-05-2026
Provider-recommended replacement: antigravity-preview-09-2026.
Google lists October 5 as the earliest shutdown date for this managed agent, rather than a base model. We have not tested whether the endpoint is already unavailable. Local-tool integrations need the new parameter and file-edit contracts described in the September 17 release notes; changing only the agent ID may be insufficient.
Read the provider notice ↗Mutable alias · snapshot updated
chat-latest
Provider-recommended replacement: Not verified.
OpenAI updated this chat snapshot on October 7. It remains a moving alias, not a newly named base model or a pinned benchmark version. The provider recommends the GPT-6 family for production APIs; no predecessor score is transferred.
Read the provider notice ↗Routed alias · now points to Haiku 5.5
~anthropic/claude-haiku-latest
Replacement recommendation: anthropic/claude-haiku-5.5.
The exact catalog alias changed from Haiku 4.5 to Haiku 5.5 between the October 7 and October 8 observations. Thinking defaults, tokenization and long-context prices change with it. Pin and test a specific model before depending on this alias; native Anthropic IDs are separate contracts.
Read the endpoint catalog ↗Deprecated; shutdown scheduled
deep-research-pro-preview-12-2025
Provider-recommended replacement: deep-research-preview-04-2026 or deep-research-max-preview-04-2026.
This applies to the managed Deep Research agent, not the underlying Gemini model family. Update the agent parameter in interactions.create and test speed versus comprehensiveness on your research tasks.
Read the provider notice ↗Gateway offer returned October 13
deepseek/deepseek-v4.1-flash-fast
Provider-recommended replacement: Not verified.
This route returned October 13 after its October 9 absence. Current output and prices differ from the previous offer; see the current profile before using it.
Read the provider notice ↗Native alias and retired-ID fallback
drex-latest / drex-v1.1
Provider-recommended replacement: drex-v2.0 / drex-v1.5.
Since October 12, omitted model and drex-latest select v2.0. Retired drex-v1.1 still selects v1.5. Current v2.0 refuses requests above its limits; the previously documented long-request fallback has been removed.
Read the provider notice ↗Native cache-price change in effect
step-5-preview
Since October 14 at 00:30 Beijing time, the standard native cache-read price increased from ¥0.35 to ¥0.70 per million tokens. Users with a pay-as-you-go call before the cutoff keep ¥0.35. Step Plan is excluded. Uncached input ¥7 and output ¥20 remain unchanged. These native yuan rates do not replace routed dollar quotes.
Read the provider notice ↗Deprecated; retirement scheduled
gpt-4o-2024-08-06
Provider-recommended replacement: gpt-6-luna.
OpenAI plans to remove this API snapshot on January 14, 2027. Fine-tuned versions follow the same schedule. Test the replacement on your own work; old benchmark scores do not transfer.
Read the provider notice ↗Deprecated; retirement scheduled
gpt-4o-2024-11-20
Provider-recommended replacement: gpt-6-luna.
OpenAI plans to remove this API snapshot on January 14, 2027. Fine-tuned versions follow the same schedule. Test the replacement on your own work; old benchmark scores do not transfer.
Read the provider notice ↗Deprecated; retirement scheduled
gpt-4o-mini-2024-07-18
Provider-recommended replacement: gpt-6-luna.
OpenAI plans to remove this API snapshot on January 14, 2027. Fine-tuned versions follow the same schedule. Test the replacement on your own work; old benchmark scores do not transfer.
Read the provider notice ↗Deprecated; retirement scheduled
gpt-4.1-2025-04-14
Provider-recommended replacement: gpt-6-luna.
OpenAI plans to remove this API snapshot on January 14, 2027. Fine-tuned versions follow the same schedule. Test the replacement on your own work; old benchmark scores do not transfer.
Read the provider notice ↗Deprecated; retirement scheduled
gpt-4.1-mini-2025-04-14
Provider-recommended replacement: gpt-6-luna.
OpenAI plans to remove this API snapshot on January 14, 2027. Fine-tuned versions follow the same schedule. Test the replacement on your own work; old benchmark scores do not transfer.
Read the provider notice ↗Preview route absent; native stable release available
unbiased/pareto-26.10-preview
Replacement recommendation: Not verified.
The October 16 OpenRouter catalog no longer lists this preview. The provider released stable 26.10 on October 15 under native ID pareto-26.10; pareto now selects it.
Read the endpoint catalog ↗Moving alias · points to Pareto 26.9
~unbiased/pareto-latest
Replacement recommendation: unbiased/pareto-26.9.
This alias currently selects Pareto 26.9. It can move to another version without changing your request ID; use a named version when repeatability matters.
Read the endpoint catalog ↗Introductory prices end November 30
gemini-4-argon
Standard input/cache/output prices per million tokens change from $2/$0.10/$10 to $4/$0.20/$20 on December 1. Batch, Flex and Priority have separate rates; thinking is billed as output.
Read the provider notice ↗Documented model ID changed
Beam-1.0-501B-A23B
Provider-recommended replacement: Not verified.
Current examples use Beam-1.0-501B-A23B instead of Beam-501B-A23B. Check the models available to your account. A new spelling is not evidence of a new model checkpoint or a retirement date.
Read the provider notice ↗Version changed
deepseek-flash
Provider-recommended replacement: DeepSeek-V4.1-Flash-1017.
Native deepseek-flash and the retired V4 Flash names now run DeepSeek-V4.1-Flash-1017. Earlier benchmark results remain attached to the versions that were tested.
Read the provider notice ↗Deprecated; automatic redirect scheduled
gemini-2.5-pro
Provider-recommended replacement: gemini-3.8-flash.
Google will automatically serve the replacement and charge its rates from December 1. Existing-user access remains until then; AI Studio selection has ended. Test tool calls, thinking settings and sampling compatibility before migration. Historical scores remain attached to the old version.
Read the provider notice ↗Deprecated; automatic redirect scheduled
gemini-2.5-flash
Provider-recommended replacement: gemini-3.5-flash-lite.
Google will automatically serve the replacement and charge its rates from December 1. Existing-user access remains until then; AI Studio selection has ended. Test tool calls, thinking settings and sampling compatibility before migration. Historical scores remain attached to the old version.
Read the provider notice ↗SAME TOKEN ASSUMPTIONS · DIFFERENT BILL
Pick two models and enter your monthly usage. The estimate uses OpenRouter offers checked 2026-11-03.
Tokens are the pieces of text a model processes. Include charged thinking tokens in output. Caching reuses previously stored input, so those tokens are already part of your total input. This comparison uses the same counts for both models; actual usage, answer quality and caching can differ.
Catalog standard token rates.
Check this offer ↗Catalog standard token rates.
Check this offer ↗Same estimate with these token assumptions.
The same text can use more tokens with a new tokenizer. Haiku 5.5 counts about 30% more than Haiku 4.5 on average, so repeat this estimate with your measured usage. Excludes cache writes, retries, tool fees, image/audio-specific billing, hosting, taxes and service tiers. The caching estimate assumes your reusable input is already stored and qualifies for a discount. Rules differ by service. Only confirmed catalog prices for long prompts are included. A large input limit does not guarantee the model will use every detail correctly.
Claude Haiku 5.5, GPT-6.1 Sol and Claude Sonnet 5.5 have extra high and max results on the same LiveBench question set. Lower settings remain unmeasured in this dataset. Their dollar costs and durations are not published here.
This page combines the last daily catalog review with individually dated new-model additions. It is not a live feed. A source checked today can still contain older results. Missing independent evidence stays pending.
Reviewed text models, serving offers, composite systems and specialized decision and research models discovered since the September 23 snapshot. These are distinct kinds of systems, not a count of independent base models. Discovery and catalog addition dates are not verified launch dates. Upgrade cost scenarios use selected OpenRouter offers; other documented prices are shown separately.
Download daily review archive ↓ · Current model benchmark snapshot ↓ · Current effort snapshot ↓