Superpower Frontier Model Ledger

Name: Superpower Frontier Model Ledger
Creator: Superpower Daily
License: https://creativecommons.org/licenses/by/4.0/

Superpower Daily

Research observatory 14 / Price, context and releases / Continuous

Superpower Frontier Model Ledger

A dated ledger of disclosed model prices, context windows, availability, and release conditions.

Archived snapshotv40Sep 24, 2026

View latest dataset

Dated observations: 107
Verified model records
Priced observations: 30
At least one disclosed token price
Availability labels: 100
Distinct disclosed conditions
Largest context: 11M
Disclosed tokens

Data status939.095 verified records across 5 periods

Each dot is a verified monthly observation of usd per million tokens. Lines connect observations in the same series; missing periods are not treated as zero.

Verified observationHover or focus any mark for exact valuesLast updated Sep 24, 2026

Version ledger

Frozen public editions

Each edition preserves the records, method, sources, and downloads available at publication time.

Coverage note

Systematic change history begins with stored observations. Historical price changes that predate collection are not reconstructed without dated primary evidence.

Dataset ID: spd:model-api-history
Stable URL: /research/model-api-history
Cadence: Continuous
Version: v40
Coverage: 2024-10-07/2026-09-23
Records: 107
Fields: 7
Updated: Sep 24, 2026
Reuse license: CC BY 4.0

Open reuse license

Reuse this data and these charts.

You may copy, publish, and adapt these selected research assets, including commercially, under CC BY 4.0 .

Credit Superpower Daily, link to the report and license, and indicate if you made changes.

This license covers Superpower Daily's original dataset compilation and charts on this page. It does not cover article text, third-party source material, logos, screenshots, photographs, or other assets unless explicitly stated.

Required attribution

Source: Superpower Daily, "Superpower Frontier Model Ledger," https://superpowerdaily.com/research/model-api-history, licensed under CC BY 4.0.

Read the data

The records behind the figure

CSV JSON

Superpower Frontier Model Ledger data records
Observed	Model	Developer	Input / 1M	Output / 1M	Availability	Context
2024-10-07	Inflection 3.0	Inflection AI	—	—	Commercial API available to developers; enterprise deployment on-premises, in the cloud, or hybrid	—
2025-08-22	GPT-4b micro	OpenAI	—	—	developed for research purposes; not broadly available	64000
2025-12-11	GPT-5.2	OpenAI	—	—	In ChatGPT: GPT-5.2 Instant, Thinking, and Pro begin rolling out today to paid plans; In the API: available now to all developers.	262144
2025-12-16	GPT Image 1.5	OpenAI	—	—	Model rolling out in ChatGPT for all users and available in the API as GPT Image 1.5; the new Images experience is rolling out to most ChatGPT users, with Business and Enterprise access coming later; global.	—
2026-04-07	Claude Mythos Preview	Anthropic	25.000000	125.000000	restricted access (restricted-access rollout)	1000000
2026-04-28	Laguna XS.2	Poolside	—	—	Preview; Hugging Face, Poolside API, and OpenRouter	—
2026-06-09	Claude Fable 5	Anthropic	10.000000	50.000000	Publicly accessible via API and Enterprise; included in subscription plans through 2026-06-22 after which access requires purchase of usage credits; some sensitive-topic access restricted and Mythos 5 limited to vetted Project Glasswing participants	—
2026-06-30	Claude Sonnet 5	Anthropic	2.000000	10.000000	All Claude plans and Claude API	—
2026-07-02	Laguna XS 2.1	Poolside	0.100000	0.200000	Poolside API, OpenRouter, Hugging Face weights, and local deployment	256000
2026-07-08	GPT-Live	OpenAI	—	—	Rolling out to ChatGPT users globally (ChatGPT Voice); API availability planned/coming soon. (Update 2026-07-31: SynthID watermarking and API verification access added for supported audio.)	—
2026-07-09	GPT-5.6 (Sol, Terra, Luna)	OpenAI	—	—	General availability following a limited preview	—
2026-07-16	Kimi K3	Moonshot	—	—	Weights published for anyone to download; model can be run locally on user hardware	1000000
2026-07-24	Claude Opus 5	Anthropic	—	—	Available immediately; default on Claude Max and strongest model on Claude Pro	—
2026-07-31	MiniMax H3	MiniMax	—	—	Hosted API globally; downloadable weights restricted to the license's applicable territories	—
2026-08-01	abliterated-model-large-v2	Abliteration.ai	5.000000	5.000000	Commercial API access	—
2026-08-10	Muse Glimmer	Meta	—	—	Consumer hardware with 24 GB or 32 GB of memory	131072
2026-08-10	Muse Glimmer	Meta Platforms	—	—	Developers can download and modify the model	—
2026-08-11	Nemotron 3.5 Lightning	Nvidia	—	—	Available through Nvidia and model repositories including Hugging Face and ModelScope; NeMo Switchyard available on GitHub	1000000
2026-08-12	Qwen3.8-2.4T-A95B	Alibaba Cloud Qwen team	—	—	Downloadable weights and configuration files for self-hosting; managed access also available through Qwen Cloud	262144
2026-08-12	Qwen3.8-27B-CRACK-GGUF	Jinho Jang (dealign.ai)	—	—	Hugging Face repository; downloadable GGUF files and a 0.9 GB vision projector for local use	262000
2026-08-13	GLM-5.3	Z.ai	1.400000	440.000000	Together AI serverless and dedicated infrastructure	1000000
2026-08-13	Gemini 3.7 Flash	Google	0.750000	3.750000	Available via Google AI Studio, Android Studio, Google Antigravity, Gemini Enterprise Agent Platform, the Gemini Enterprise app, and Gemini Spark for Google AI Pro and Ultra subscribers in supported countries.	—
2026-08-16	GLM-5.3	Zhipu (Z.ai)	—	—	—	—
2026-08-18	GPT-5.6 Sol	OpenAI	—	—	Limited release to a small group of customers; wider access planned later	—
2026-08-19	Ornith-1.5	Ornith AI	—	—	Hugging Face collection; local deployment and quantized mobile version described	—
2026-08-21	Falcon 2	Murf AI	—	—	—	—
2026-08-21	V4-Flash-Vision-Exp	DeepSeek	—	—	DeepSeek API	—
2026-08-21	deepseek-v4-flash-vision-exp	DeepSeek	0.220000	0.660000	Available through Chat Completions, Messages, and Responses requests	—
2026-08-24	HiDream-O1-World	HiDream.ai	—	—	—	—
2026-08-24	Thomson-1.0-Large	Thomson Reuters	—	—	—	—
2026-08-24	Thomson	Thomson Reuters	—	—	Initially available in CoCounsel Legal’s Tabular Analysis feature; smaller version planned for Hugging Face	—
2026-08-24	GEN-1.5	Generalist AI	—	—	Available through direct partnerships on Generalist AI's own robot fleet; no public weights, API, or self-serve product	—
2026-08-26	Grok 4.6	xAI	2.000000	6.000000	Microsoft Foundry for Azure customers	500000
2026-08-26	Granite 4.2	IBM	—	—	Hugging Face, Ollama, GitHub, and other platforms	512000
2026-08-26	Qwen3.8-Flash-Next	Alibaba Qwen team	0.160000	0.470000	Weights on Hugging Face and ModelScope; production Qwen3.8-Flash through QwenCloud; API expected shortly	262144
2026-08-26	Qwen3.8-Flash	Alibaba	1.000000	3.000000	API access	262144
2026-08-26	Qwen3.8-Flash	Alibaba Qwen	0.160000	0.470000	QwenCloud managed API	1000000
2026-08-27	Gemini 3.5 Transcribe	Google	—	—	Public preview	—
2026-08-27	GLM-5.3-Flash	Z.ai	0.150000	0.500000	Z.ai API; weights available on Hugging Face	1000000
2026-08-27	GLM-5.3-Flash	Zhipu AI	0.150000	0.500000	Officially released; deployable on domestically produced AI chips	1000000
2026-08-27	leiolai-1	Leiolai	0.010000	0.020000	Consumer app and developer API	11000000
2026-08-28	ABot-Recon	AMAP CV Lab	—	—	Inference code, model checkpoint, and evaluation code released; weights limited to noncommercial research and education	—
2026-08-28	Sahara v2.5	Intron	—	—	—	—
2026-08-31	GigaPath-Flash and GigaTIME-Flash	Microsoft Research	—	—	Hugging Face	—
2026-08-31	S1	Skild AI	—	—	—	—
2026-08-31	Isaac 0.5	Perceptron AI	—	—	Model weights available through Hugging Face; fine-tuning and inference code available through GitHub	—
2026-09-01	Orbis 1.0	Visko	—	—	public access	—
2026-09-01	Solaris	Runway	—	—	Early access via application form; launch partners sought	—
2026-09-01	MAPL-EMIT	Google Research	—	—	Trained model and synthetic plume dataset on Kaggle; global plume database on Earth Engine; inference library on GitHub	—
2026-09-01	Claude Fable 5.1 and Claude Mythos 5.1	Anthropic	10.000000	50.000000	Claude Fable 5.1 is generally available through the Claude API, Amazon Bedrock, Claude Platform on AWS, Google Cloud, and Microsoft Foundry; Claude Mythos 5.1 is restricted to vetted U.S. organizations in Project Glasswing.	1000000
2026-09-01	Muse Voice Transcribe	Meta Superintelligence Labs	—	—	Meta AI for Mac, Muse Code, and Meta Model API	—
2026-09-02	Muse Spark 1.3	Meta	1.250000	4.250000	Broad rollout through Muse Code and Meta Model API for xhigh; max configuration limited to partner preview pending additional safety testing	—
2026-09-02	Muse Spark 1.3	Meta AI Research	—	—	Rolling out in Muse Code and Meta Model API	—
2026-09-02	Gemini 3.8 Flash; Gemini 3.8 Flash Cyber	Google	0.750000	3.750000	Gemini 3.8 Flash available through Google AI Studio and Android Studio; Gemini 3.8 Flash Cyber restricted to trusted governmental and critical-infrastructure entities through the Fairwind Program	1048576
2026-09-02	Gemini 3.8 Flash	Google	0.750000	3.750000	Gemini API, Google AI Studio, Android Studio, Google Antigravity, Gemini Enterprise, Gemini app, AI Mode in Google Search, and Gemini in Google Sheets	1000000
2026-09-02	Gemini 3.8 Flash and Gemini 3.8 Flash Cyber	Google	0.750000	3.750000	Gemini 3.8 Flash is available in Gemini Enterprise and through Google AI Studio, Google Antigravity, Android Studio, and Stitch; Gemini 3.8 Flash Cyber is limited to trusted defenders through the Fairwind Program.	1000000
2026-09-03	K2 Horizon	Institute of Foundation Models	—	—	Hugging Face repositories	—
2026-09-03	WeatherNext 3	Google DeepMind and Google Research	—	—	Global availability across Google Search, Gemini, Google Maps, Google Maps Platform Weather API, Google Earth Engine, BigQuery, and Google Cloud Storage	—
2026-09-03	humain-m3	HUMAIN	—	—	HUMAIN Node; limited-preview and research-preview tiers	—
2026-09-04	GPT-6	OpenAI	—	—	Select cybersecurity customers initially; wider rollout to paying customers; unavailable to free-tier and cheapest-plan users	—
2026-09-06	Lyria 3.5	Google	—	—	Gemini app, Google Flow Music, Google AI Studio, and Google Vids	—
2026-09-06	WeatherNext 3	Google DeepMind	—	—	Google Search, Gemini, Google Maps, Google Maps Platform Weather API, Google Earth Engine, BigQuery, and bulk download via Google Cloud Storage	—
2026-09-06	Muse Voice Transcribe	Meta	—	—	Meta AI, Muse Code, and Meta Model API	—
2026-09-07	Qwen-Drive 1.0	Alibaba	—	—	Free to the research community via Hugging Face, ModelScope, and GitHub	—
2026-09-07	Astra	OpenAI	—	—	Generally available	—
2026-09-08	OUI-1	Thesys	—	—	Local self-hosting; Thesys says it runs at FP8 on an Nvidia RTX 5090	—
2026-09-08	N2.5 agent family	Nex-AGI	—	—	Mini and Max available; Pro weights marked coming soon	—
2026-09-09	Suno v6	Suno	—	—	v6, v6-wild, and free tier v6-mini	—
2026-09-09	V6	Suno	—	—	Available through three new subscription tiers	—
2026-09-09	Granite PatchTST-FM-r2	IBM	—	—	Downloadable from the Hugging Face Hub; available through granite-tsfm 0.3.9 or later	8192
2026-09-09	GPT-Image-2.5 Flare and GPT-Image-2.5 Sunburst	OpenAI	8.000000	30.000000	Worldwide availability in ChatGPT, ChatGPT Work, Codex, and the API	—
2026-09-10	Adaptive Instructed-Retriever	Databricks	—	—	—	—
2026-09-10	DeepSeek-V4.1-Flash	DeepSeek	0.300000	1.200000	DeepSeek first-party API; US/API availability reported; also supported by Baseten and Ollama	1000000
2026-09-10	NASA-IBM Lunar Foundation Model	NASA and IBM	—	—	Hugging Face weights; GitHub fine-tuning and inference code	—
2026-09-10	Edge0 35B and Edge0 8B	Edge0	—	—	Released for macOS on Apple Silicon; no iOS build published	—
2026-09-10	Aether	JoyIn	—	—	Publicly announced; selected components planned for open-source release	—
2026-09-10	GPT-Live-1	OpenAI	—	—	Available to developers	—
2026-09-11	GPT-6 Astra	OpenAI	10.000000	50.000000	ChatGPT Plus, Pro, Business, and Enterprise; OpenAI API	1050000
2026-09-13	Music v2.5	ElevenLabs	—	—	ElevenMusic app and API; free and Pro tiers	—
2026-09-13	Iris-mini and Iris-pro	AllSpark	—	—	Model weights on Hugging Face; code and Iris Harness on GitHub	256000
2026-09-15	Boreal	Creatify Labs	—	—	Available now through the Creatify Labs API, Creatify Model Playground, AdFlow features, and fal platform	—
2026-09-15	Gemini 3.8 Live	Google DeepMind	—	—	Available through the Gemini API, Google AI Studio, and Live API integrations	—
2026-09-15	Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking	Google DeepMind	—	—	Gemini API, Gemini app, Google AI Studio, Google Cloud, Google Search Live, and selected Google Workspace surfaces; Live API documentation remains labeled preview	131072
2026-09-15	Gemini 3.8 Live; Gemini 3.8 Live Extended Thinking	Google DeepMind	—	—	Available to developers through the Gemini API and Google AI Studio	—
2026-09-15	TabPFN-3.5	Prior Labs	—	—	Open weights available for research, evaluation, and Kaggle; production use through Prior Labs API or a commercial license	—
2026-09-15	Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking	Google	3.000000	12.000000	Available in the Gemini Live API and Google AI Studio for production use	—
2026-09-16	Pareto 26.9	unbiased.ai	2.500000	7.500000	OpenRouter and Cloudflare	262144
2026-09-16	Koa	Salesforce	—	—	Piloting with Formula 1, Baxter Credit Union, and Xero; general availability planned for winter	—
2026-09-18	Grok Voice Transcribe 2.0	xAI	—	—	Available through REST and WebSocket Speech-to-Text API endpoints; selectable model, with planned default status	—
2026-09-18	Bonsai 2 27B	PrismML	—	—	Weights available at launch; runs on Nvidia GPUs via CUDA and Apple devices via MLX	—
2026-09-18	Ternary Bonsai 2 27B	PrismML	—	—	Released; local deployment via PrismML’s llama.cpp fork or MLX runtime	262000
2026-09-19	Qwen3.8-Omni-Flash	Qwen	0.150000	0.470000	Available through Qwen Studio, Qwen Cloud, and the Qwen API	1000000
2026-09-20	Qwen-Image-2.1	Alibaba Qwen AI team	—	—	Available on Hugging Face, GitHub, Model Scope, and via a Hugging Face demo	—
2026-09-21	Alice AI Foundation	Yandex	—	—	Commercial and research use	—
2026-09-21	Altar	Aikido	—	—	Available openly for local deployments	—
2026-09-21	RetroChimera	Microsoft Research	—	—	GitHub and Microsoft Foundry	—
2026-09-21	Qwen-Image-2.1	Alibaba Qwen	—	—	Research and evaluation; commercial deployment requires a separate Qwen license	—
2026-09-21	Grok 4.7	SpaceXAI	2.000000	6.000000	Available in the Grok API, Cursor, Grok Build, third-party coding tools, model routers, and cloud platforms; gradual GitHub Copilot rollout	500000
2026-09-21	Grok 4.7	SpaceX	2.000000	6.000000	Available from $2 per million input tokens and $6 per million output tokens	—
2026-09-22	Xing4.0-29B-A4B	China Telecom AI	—	—	GitHub and Hugging Face repositories	256000
2026-09-22	MiMo-V2.6-Pro	Xiaomi	0.435000	0.870000	Weights downloadable from Xiaomi's Hugging Face repository; API access listed through OpenRouter and other resellers.	1000000
2026-09-22	GPT-6 Sol and GPT-6 Luna	OpenAI	—	—	API; ChatGPT Work and Codex for Plus, Pro, Business, Enterprise, and Edu users; Luna also available in the ChatGPT desktop app for Free and Go subscribers	1050000
2026-09-22	Claude Opus 5.5	Anthropic	4.000000	20.000000	Anthropic API, AWS, Google Cloud, and Microsoft’s ecosystem	1000000
2026-09-23	Nemotron 3 Diarization	NVIDIA	—	—	Available on Hugging Face	—
2026-09-23	FLUX 3 Action	Black Forest Labs	—	—	Previously accessible through selected research and commercial partners; public weights were not yet available according to the article	—
2026-09-23	Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS	Google	—	—	Rolling out through the Gemini API and Google AI Studio; enterprise API access via Gemini Enterprise is listed as coming soon.	—
2026-09-23	Opus 5.5	Anthropic	—	—	Launched for Claude; special permission required for cyber security and biology researchers using it in their work	—

Measurement technique

How to read this report

01Each row is a dated verified model record; later records can preserve subsequent price or availability changes.
02Prices are shown exactly as disclosed per million tokens and are not adjusted for caching, batching, or enterprise discounts.
03The chart is an observation history, not a claim that unchanged prices persisted between observations.

Sources

Evidence

26 publishers supporting 41 records. Expand a publisher to inspect its cited pages.

the-decoder.comthe-decoder.com6 sources / 6 records

Thomson Reuters bets $40M on owning its AI instead of renting from OpenAI or Anthropic1 supporting record IBM drops open-weight Granite 4.2 family with built-in agentic capabilities under Apache 2.01 supporting record Alibaba releases Qwen3.8-Flash-Next, targeting "ultimate cost efficiency"1 supporting record Google's Gemini 3.5 Transcribe turns speech to text in 85 languages while auto-correcting your verbal stumbles1 supporting record GLM-5.3-Flash matches top models at a fraction of the cost, and runs without Nvidia1 supporting record Runway's Solaris is an AI system that generates software interfaces in real time1 supporting record

Ars Technicaarstechnica.com3 sources / 3 records

IBM's new Granite 4.2 models ride the wave of interest in local LLMs1 supporting record Google releases Gemini 3.8 Flash, its third Flash model in six weeks1 supporting record Google announces Gemini 3.5 Transcribe for AI-powered speech-to-text1 supporting record

marktechpost.commarktechpost.com3 sources / 3 records

Generalist AI Releases GEN-1.5: A Robot Foundation Model That Learns New Tasks From One 3–12 Second Demo1 supporting record Alibaba's Qwen Team Releases Qwen3.8-Flash-Next: A 125B Multimodal MoE With 6B Active Parameters Previewing the Qwen4 Architecture1 supporting record Z.ai Releases GLM-5.3-Flash: A 320B-A18B Natively Multimodal MoE With a 1M-Token Context1 supporting record

openai.comopenai.com2 sources / 3 records

GPT-6 Astra: A new generation of intelligence2 supporting records GPT-5.6: Frontier intelligence that scales with your ambition1 supporting record

blog.googleblog.google2 sources / 2 records

Intelligent transcription with Gemini 3.5 Transcribe1 supporting record Introducing Gemini 3.7 Flash1 supporting record

campustechnology.comcampustechnology.com2 sources / 2 records

Meta Releases Muse Glimmer, an Open-Weight AI Model Designed to Run on Consumer Hardware -- Campus Technology1 supporting record Meta Releases Muse Glimmer, an Open-Weight AI Model Designed to Run on Consumer Hardware -- Campus Technology1 supporting record

Show 20 more publishers

siliconangle.comsiliconangle.com2 sources / 2 records

Thomson Reuters launches proprietary AI model for legal work - SiliconANGLE1 supporting record Z.ai open-sources ‘Ox Alpha’ model as GLM-5.3-Flash - SiliconANGLE1 supporting record

x.aix.ai2 sources / 2 records

Introducing Grok 4.61 supporting record Grok 4.6 on Amazon Bedrock1 supporting record

aithority.comaithority.com1 source / 1 record

Perceptron AI Launches Isaac 0.5, a Frontier Open-Weight Robotics Model1 supporting record

autoclaw.z.aiautoclaw.z.ai1 source / 1 record

AutoClaw Models | GLM Model Support1 supporting record

CNBCcnbc.com1 source / 1 record

Z.ai shares surge 8% after releasing new AI model running only on Chinese chips1 supporting record

cryptobriefing.comcryptobriefing.com1 source / 1 record

Zhipu launches GLM-5.3-Flash, its first natively multimodal model built for Chinese chips1 supporting record

deccanchronicle.comdeccanchronicle.com1 source / 1 record

Alibaba Releases Smaller, Cost-Effective Qwen AI Model1 supporting record

economictimes.comeconomictimes.com1 source / 1 record

Alibaba's Qwen launches Qwen3.8-Flash AI model with lower training costs - The Economic Times1 supporting record

fool.comfool.com1 source / 1 record

Mark Zuckerberg's Meta Just Open-Sourced Its Most Powerful AI Model to Take on OpenAI and Anthropic. Should Investors Watch Meta's AI Spending Closely? - Yahoo Finance1 supporting record

globaltimes.cnglobaltimes.cn1 source / 1 record

Zhipu AI releases GLM-5.3-Flash model, powered by 100,000 domestically made chips - Global Times1 supporting record

invezz.cominvezz.com1 source / 1 record

Alibaba unveils cheaper Qwen AI model: is this what its stock needed?1 supporting record

itbrief.newsitbrief.news1 source / 1 record

Leiolai launches AI model that runs on user devices1 supporting record

microsoft.commicrosoft.com1 source / 1 record

Making pathology foundation models practical at scale1 supporting record

readthepeak.comreadthepeak.com1 source / 1 record

Thomson Reuters Launches Its Own AI Model1 supporting record

stocktitan.netstocktitan.net1 source / 1 record

Thomson Reuters launches its own AI model for legal document reviews1 supporting record

theaiinsider.techtheaiinsider.tech1 source / 1 record

Perceptron AI Launches Open-Weight Robotics Model Called 'Isaac 0.5'1 supporting record

thelogic.cothelogic.co1 source / 1 record

Thomson Reuters launches its own AI model to reduce reliance on big tech - The Logic1 supporting record

venturebeat.comventurebeat.com1 source / 1 record

DeepSeek-V4.1-Flash debuts with $0.003/1M off-peak cached-input rate and benchmarks eclipsing GPT-5.6 Sol, Claude Opus 51 supporting record

vir.com.vnvir.com.vn1 source / 1 record

HiDream.ai releases HiDream-O1-World interactive artificial intelligence model1 supporting record

yahoo.comyahoo.com1 source / 1 record

Alibaba's Qwen launches Qwen3.8-Flash AI model with lower training costs1 supporting record

Next report / 15AI Workforce Impact Monitor All research reports

Superpower Daily

Your reading space

Superpower Frontier Model Ledger

Frozen public editions

Reuse this data and these charts.

The records behind the figure

How to read this report

Evidence

Notifications