Skip to main content
Framemode
Plans
中文

A software factory.

Bootstrapped. Agent-first. Targeting niche arbitrage opportunities with B2B and B2C SaaS and agent services.

Each product has its own market, pricing, and reason to exist.

/ freɪ-moʊd /About Framemode
01 / WHY THIS MODEL

The economics
have changed.

Intelligence costs less to run. Getting a product into someone’s hands still takes work. The opening is a specific job, a reachable buyer, and less software between them.

280+×lower inference price

A different cost base.

A GPT-3.5-level benchmark score went from $20 to $0.07 per million tokens. That changes which small, repeated jobs can make sense as products.

FACTORY THESIS / Keep the scope narrow. Price the completed job.

GPT-3.5-level inference price

USD per million tokens
GPT-3.5-level inference priceNov 2022: $20 per million tokens. Oct 2024: $0.07 per million tokens. Linear scale from zero.$0$5$10$15$20Nov 2022$20.00Oct 2024$0.07
MMLU score of at least 64.8 · Nov 2022 → Oct 2024Stanford AI Index 2025 ↗

Two reported endpoints at a fixed benchmark threshold. Different models; inference price only.

02 / THE PRODUCT PRINCIPLE

Built for agents.
Usable by people.

Agent-first means the product exposes actions an agent can actually use. People get an interface. Agents get a direct route to the same job. Both need permissions, clear results, and a way to recover when something fails.

PRODUCT DESIGN MODELTwo ways into the same productA person uses the interface. An agent uses defined actions. Both pass through permissions to the product workflow, which returns an inspectable result.PERSONInterfaceAGENTActionsPRODUCTDo the jobRESULTInspect / actPERMISSIONS

Software-as-a-Service gives access to the tool.
Agents-as-a-Service takes on the task.

THE INTERFACES ARE ALREADY HERE97M+

monthly MCP SDK downloads
Python + TypeScript

10,000+active public MCP servers
Anthropic · MCP ecosystem update ↗Reported 9 Dec 2025. Python and TypeScript SDK downloads per month. Downloads are not unique users. Public servers are an ecosystem count.
03 / THE RESEARCH DESK

What intelligence costs.

Follow the published prices. Compare the companies. Inspect the evidence behind a forecast.

THE COST OF INTELLIGENCE

The market, in motion.

AUG 20221 / 120 QUOTES
USD / 1M TOKENS · LOG SCALE
94 distinct model labels from nine providers. USD per million input tokens, logarithmic vertical scale. Observations from August 2022 to November 2026. Each provider has a distinct color and can be hidden independently. Dotted curves extrapolate eligible successor-family price trends through November 2028. The selected family has a shaded sensitivity range, not a probability interval.$100$10$1$0.1$0.01202320242025202620272028FORECAST / ANNOUNCEDOpenAI: GPT flagship line. Estimated 2028 input price $1.53 per million tokens. Fitted to 6 dated observations.Anthropic: Claude Opus line. Estimated 2028 input price $3.67 per million tokens. Fitted to 5 dated observations.Google: Gemini Flash line. Estimated 2028 input price $0.10 per million tokens. Fitted to 3 dated observations.Mistral: Mistral Small line. Estimated 2028 input price $0.15 per million tokens. Fitted to 3 dated observations.DeepSeek: DeepSeek general / Pro line. Estimated 2028 input price $1.92 per million tokens. Fitted to 6 dated observations.xAI: Grok general-model line. Estimated 2028 input price $2.00 per million tokens. Fitted to 4 dated observations.Cohere: Command R efficiency line. Estimated 2028 input price $0.0368 per million tokens. Fitted to 3 dated observations.Amazon: Nova Micro · AWS us-east-1. Estimated 2028 input price $0.035 per million tokens. Fitted to 5 dated observations.Alibaba: Qwen Plus · Singapore / International. Estimated 2028 input price $0.40 per million tokens. Fitted to 2 dated observations.OpenAI / GPT-3 Ada / Aug 22, 2022 / $0.80 per 1M input tokensOpenAI / GPT-3 Babbage / Aug 22, 2022 / $1.20 per 1M input tokensOpenAI / GPT-3 Curie / Aug 22, 2022 / $6.00 per 1M input tokensOpenAI / GPT-3 Davinci / Aug 22, 2022 / $60.00 per 1M input tokensOpenAI / GPT-3 Ada / Sep 1, 2022 / $0.40 per 1M input tokensOpenAI / GPT-3 Babbage / Sep 1, 2022 / $0.50 per 1M input tokensOpenAI / GPT-3 Curie / Sep 1, 2022 / $2.00 per 1M input tokensOpenAI / GPT-3 Davinci / Sep 1, 2022 / $20.00 per 1M input tokensOpenAI / GPT-3.5 Turbo / Mar 1, 2023 / $2.00 per 1M input tokensAnthropic / Claude 2 / Jul 2023 / $11.02 per 1M input tokensAnthropic / Claude Instant / Jul 2023 / $1.63 per 1M input tokensOpenAI / GPT-3.5 Turbo 1106 / Nov 6, 2023 / $1.00 per 1M input tokensOpenAI / GPT-4 / Nov 6, 2023 / $30.00 per 1M input tokensOpenAI / GPT-4 Turbo / Nov 6, 2023 / $10.00 per 1M input tokensAnthropic / Claude 3 Opus / Mar 4, 2024 / $15.00 per 1M input tokensAnthropic / Claude 3 Sonnet / Mar 4, 2024 / $3.00 per 1M input tokensAnthropic / Claude 3 Haiku / Mar 13, 2024 / $0.25 per 1M input tokensAnthropic / Claude 3.5 Sonnet / Jun 21, 2024 / $3.00 per 1M input tokensOpenAI / GPT-4o mini / Jul 18, 2024 / $0.15 per 1M input tokensGoogle / Gemini 1.5 Flash / Aug 12, 2024 / $0.075 per 1M input tokensCohere / Command R+ 08-2024 / Aug 30, 2024 / $2.50 per 1M input tokensCohere / Command R 08-2024 / Aug 30, 2024 / $0.15 per 1M input tokensMistral / Codestral / Sep 17, 2024 / $0.20 per 1M input tokensMistral / Mistral Large 2 / Sep 17, 2024 / $2.00 per 1M input tokensMistral / Mistral NeMo / Sep 17, 2024 / $0.15 per 1M input tokensMistral / Mistral Small 24.09 / Sep 17, 2024 / $0.20 per 1M input tokensGoogle / Gemini 1.5 Flash-8B / Oct 14, 2024 / $0.0375 per 1M input tokensMistral / Ministral 3B / Oct 16, 2024 / $0.04 per 1M input tokensMistral / Ministral 8B / Oct 16, 2024 / $0.10 per 1M input tokensxAI / Grok-2 1212 / Dec 12, 2024 / $2.00 per 1M input tokensAmazon / Nova Lite / Jan 9, 2025 / $0.06 per 1M input tokensAmazon / Nova Micro / Jan 9, 2025 / $0.035 per 1M input tokensAmazon / Nova Pro / Jan 9, 2025 / $0.80 per 1M input tokensDeepSeek / DeepSeek R1 / Jan 20, 2025 / $0.55 per 1M input tokensDeepSeek / DeepSeek V3 / Feb 8, 2025 / $0.27 per 1M input tokensAmazon / Nova Lite / Apr 4, 2025 / $0.06 per 1M input tokensAmazon / Nova Micro / Apr 4, 2025 / $0.035 per 1M input tokensOpenAI / GPT-4.1 / Apr 14, 2025 / $2.00 per 1M input tokensOpenAI / GPT-4.1 mini / Apr 14, 2025 / $0.40 per 1M input tokensOpenAI / GPT-4.1 nano / Apr 14, 2025 / $0.10 per 1M input tokensAmazon / Nova Lite / May 2, 2025 / $0.06 per 1M input tokensAmazon / Nova Micro / May 2, 2025 / $0.035 per 1M input tokensAmazon / Nova Pro / May 2, 2025 / $0.80 per 1M input tokensAnthropic / Claude Opus 4 / May 22, 2025 / $15.00 per 1M input tokensAnthropic / Claude Sonnet 4 / May 22, 2025 / $3.00 per 1M input tokensMistral / Mistral Small 3.1 / Jul 10, 2025 / $0.10 per 1M input tokensAlibaba / Qwen-Plus (non-thinking) / Jul 14, 2025 / $0.40 per 1M input tokensOpenAI / GPT-5 / Aug 7, 2025 / $1.25 per 1M input tokensOpenAI / GPT-5 mini / Aug 7, 2025 / $0.25 per 1M input tokensOpenAI / GPT-5 nano / Aug 7, 2025 / $0.05 per 1M input tokensDeepSeek / DeepSeek V3.1 / Sep 5, 2025 / $0.56 per 1M input tokensxAI / Grok 4 Fast / Sep 19, 2025 / $0.20 per 1M input tokensDeepSeek / DeepSeek V3.2-Exp / Sep 29, 2025 / $0.28 per 1M input tokensAnthropic / Claude Opus 4.5 / Nov 24, 2025 / $5.00 per 1M input tokensCohere / Command R+ 08-2024 / Dec 18, 2025 / $2.50 per 1M input tokensCohere / Command R 08-2024 / Dec 18, 2025 / $0.15 per 1M input tokensCohere / Command R7B 12-2024 / Dec 18, 2025 / $0.0375 per 1M input tokensAmazon / Nova Lite / Dec 23, 2025 / $0.06 per 1M input tokensAmazon / Nova Micro / Dec 23, 2025 / $0.035 per 1M input tokensAmazon / Nova Pro / Dec 23, 2025 / $0.80 per 1M input tokensDeepSeek / DeepSeek V4 Flash Preview / Apr 24, 2026 / $0.14 per 1M input tokensDeepSeek / DeepSeek V4 Pro Preview / Apr 24, 2026 / $1.74 per 1M input tokensAmazon / Nova Lite / May 4, 2026 / $0.06 per 1M input tokensAmazon / Nova Micro / May 4, 2026 / $0.035 per 1M input tokensAmazon / Nova Pro / May 4, 2026 / $0.80 per 1M input tokensxAI / Grok 4.5 / Jul 16, 2026 / $2.00 per 1M input tokensDeepSeek / DeepSeek V4 Flash 0731 / Aug 16, 2026 / $0.44 per 1M input tokensDeepSeek / DeepSeek V4 Pro 0813 / Aug 16, 2026 / $1.32 per 1M input tokensDeepSeek / DeepSeek V4.1 Flash / Sep 10, 2026 / $0.30 per 1M input tokensxAI / Grok 4.7 / Sep 21, 2026 / $2.00 per 1M input tokensMistral / Codestral / Nov 9, 2026 / $0.30 per 1M input tokensCohere / Command R 08-2024 / Nov 9, 2026 / $0.15 per 1M input tokensCohere / Command R7B 12-2024 / Nov 9, 2026 / $0.0375 per 1M input tokensCohere / Command R 03-2024 / Nov 9, 2026 / $0.50 per 1M input tokensCohere / Command R+ 04-2024 / Nov 9, 2026 / $3.00 per 1M input tokensCohere / Command R+ 08-2024 / Nov 9, 2026 / $2.50 per 1M input tokensDeepSeek / DeepSeek V4.1 Flash 1017 / Nov 9, 2026 / $0.30 per 1M input tokensDeepSeek / DeepSeek V4 Pro 0813 / Nov 9, 2026 / $1.32 per 1M input tokensAnthropic / Claude Fable 5.1 / Nov 9, 2026 / $10.00 per 1M input tokensGoogle / Gemini 2.5 Flash-Lite / Nov 9, 2026 / $0.10 per 1M input tokensGoogle / Gemini 2.5 Pro / Nov 9, 2026 / $1.25 per 1M input tokensGoogle / Gemini 3 Flash Preview / Nov 9, 2026 / $0.50 per 1M input tokensGoogle / Gemini 3.1 Pro Preview / Nov 9, 2026 / $2.00 per 1M input tokensGoogle / Gemini 3.6 Flash / Nov 9, 2026 / $0.75 per 1M input tokensGoogle / Gemini 3.8 Flash / Nov 9, 2026 / $0.75 per 1M input tokensOpenAI / GPT-4o / Nov 9, 2026 / $2.50 per 1M input tokensOpenAI / GPT-5.1 / Nov 9, 2026 / $1.25 per 1M input tokensOpenAI / GPT-5.2 / Nov 9, 2026 / $1.75 per 1M input tokensOpenAI / GPT-5.4 mini / Nov 9, 2026 / $0.75 per 1M input tokensOpenAI / GPT-5.4 nano / Nov 9, 2026 / $0.20 per 1M input tokensOpenAI / GPT-5.4 / Nov 9, 2026 / $2.50 per 1M input tokensOpenAI / GPT-5.5 / Nov 9, 2026 / $5.00 per 1M input tokensOpenAI / GPT-5.6 Luna / Nov 9, 2026 / $0.20 per 1M input tokensOpenAI / GPT-5.6 Sol / Nov 9, 2026 / $4.00 per 1M input tokensOpenAI / GPT-5.6 Terra / Nov 9, 2026 / $2.00 per 1M input tokensOpenAI / GPT-6 Astra / Nov 9, 2026 / $10.00 per 1M input tokensOpenAI / GPT-6 Luna / Nov 9, 2026 / $0.10 per 1M input tokensOpenAI / GPT-6 Sol / Nov 9, 2026 / $2.00 per 1M input tokensOpenAI / GPT-6.1 Sol / Nov 9, 2026 / $2.00 per 1M input tokensOpenAI / GPT-6.1 Terra / Nov 9, 2026 / $1.00 per 1M input tokensxAI / Grok 4.7 / Nov 9, 2026 / $2.00 per 1M input tokensAnthropic / Claude Haiku 4.5 / Nov 9, 2026 / $1.00 per 1M input tokensAnthropic / Claude Haiku 5.5 / Nov 9, 2026 / $0.10 per 1M input tokensAmazon / Nova Lite / Nov 9, 2026 / $0.06 per 1M input tokensAmazon / Nova Pro / Nov 9, 2026 / $0.80 per 1M input tokensAmazon / Nova Micro / Nov 9, 2026 / $0.035 per 1M input tokensMistral / Ministral 3 14B / Nov 9, 2026 / $0.20 per 1M input tokensMistral / Ministral 3 3B / Nov 9, 2026 / $0.10 per 1M input tokensMistral / Ministral 3 8B / Nov 9, 2026 / $0.15 per 1M input tokensMistral / Mistral Small 4 / Nov 9, 2026 / $0.15 per 1M input tokensMistral / Mistral Large 4 / Nov 9, 2026 / $0.70 per 1M input tokensOpenAI / o3 / Nov 9, 2026 / $2.00 per 1M input tokensAnthropic / Claude Opus 4.6 / Nov 9, 2026 / $5.00 per 1M input tokensAnthropic / Claude Opus 4.7 / Nov 9, 2026 / $5.00 per 1M input tokensAnthropic / Claude Opus 4.8 / Nov 9, 2026 / $5.00 per 1M input tokensAnthropic / Claude Opus 5 / Nov 9, 2026 / $5.00 per 1M input tokensAnthropic / Claude Opus 5.5 / Nov 9, 2026 / $4.00 per 1M input tokensAlibaba / Qwen3.7-Plus / Nov 9, 2026 / $0.40 per 1M input tokensAnthropic / Claude Sonnet 4.6 / Nov 9, 2026 / $3.00 per 1M input tokensAnthropic / Claude Sonnet 5.5 / Nov 9, 2026 / $2.00 per 1M input tokensGoogle / Gemini 3.8 Flash / Jan 1, 2027 / $1.50 per 1M input tokensGoogle / Gemini 3.6 Flash / Jan 1, 2027 / $1.50 per 1M input tokens

Swipe to explore the chart →

NEWS / SOURCES
○ Observed– – Announced··· ForecastSTABLE = unchanged recordBASELINE = short history–· Compute stress
2028
2020–2022 / THE API ERA

A model becomes a building block.

GPT-3 reached the OpenAI API in June 2020. The first dated price sheet plotted here is from August 2022: Davinci at $60 per million tokens.

2028 / ENSEMBLE$1.53 / 1M input

6 dates · 4 retrospective splits · weighted model ensemble
Model range: $0.73–$2.00. Not a confidence interval. Compute-stress value: $1.53.

9 / 9 COMPANIES · TREND / STABLE / BASELINESOURCE STATUS / 2026-11-09 ↗
280×+

The fixed-capability result still matters: Stanford tracked a GPT-3.5-level MMLU score from $20 to $0.07 per million tokens, Nov 2022 to Oct 2024.

Stanford AI Index 2025 ↗
Not every line falls.

Google’s published Gemini 3.8 Flash rates rise from $0.75 / $3.75 to $1.50 / $7.50 when introductory pricing ends on 1 Jan 2027. That announced change is separate from the fitted forecasts.

Google’s published schedule ↗
Prices, sources & comparison rules

A curated history of major text-model families, not every model or hosting offer. Rates are USD per million tokens, standard uncached short-context pricing. Model tokenizers, quality, context tiers, availability, and reasoning-token use differ. Low token prices do not establish a lower cost per successful job. Lines connect selected releases within families; they do not imply continuous availability or an unchanged model. Current catalog snapshots use the observation date, not a guessed release date.

How the forecasts are calculated

Forecasts combine an unchanged-price baseline, an exponentially damped log-linear trend and a robust median-slope trend. Trends need three distinct dates spanning 18 months. Weights use chronological, rolling-origin absolute log errors and shrink toward unchanged pricing when validation is sparse. The damping time is an explicit 18-month assumption. The range shows model disagreement, not a confidence interval. Sparse series use persistence with no invented uncertainty. These curated historical comparisons have selection bias; they do not establish future accuracy or equal model capability.

News and compute covariates are recorded with first-seen timestamps. A regularized cost/news regression is enabled only with 18 complete observations across a year, and is evaluated using past features carried forward, never future observed features. The new collector has insufficient history for that model today. The news row links to published source events; it does not assert a causal effect on prices. Supply, demand, energy and efficiency controls add a separate structural stress overlay; their elasticities are assumptions, not learned facts.

The chart draws one family per company, prioritizing an eligible family with the most dated history, then the longest span. A newer cheap model does not displace a better-supported forecast. A different family can be selected. Direct vendor and OpenRouter routes remain separate series. A source receipt alone does not mean its numeric rows were verified; the inspector and source-health section distinguish current row matches, parsed historical archives, dated announcements, manual observations, failures and stale sources.

For context, Epoch AI’s 2026 research ↗ measures cost at a fixed level of benchmark performance. Its market-wide results are not used as a company’s future token-price discount.

Forecast diagnostics · selected price unit · no calibrated confidence claims
Family / basisDates / splitsMethod log errors2028 estimateModel range
OpenAI
GPT flagship linehistorical ensemble · Anchor: GPT-6.1 Sol
6 / 4Aug 22, 2022 – Nov 9, 2026persistence: 0.811damped-trend: 8.575robust-trend: 0.744$1.53$0.73 – $2.00
OpenAI
GPT small-model linehistorical ensemble · Anchor: GPT-6 Luna
6 / 4Mar 1, 2023 – Nov 9, 2026persistence: 0.922damped-trend: 1.069robust-trend: 1.076$0.0707$0.0316 – $0.10
OpenAI
GPT-5.4 nanopersistence baseline · Anchor: GPT-5.4 nano
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$0.20Not calibrated
OpenAI
GPT-5.6 Lunapersistence baseline · Anchor: GPT-5.6 Luna
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$0.20Not calibrated
OpenAI
GPT-5.4 minipersistence baseline · Anchor: GPT-5.4 mini
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$0.75Not calibrated
OpenAI
GPT-6.1 Terrapersistence baseline · Anchor: GPT-6.1 Terra
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$1.00Not calibrated
OpenAI
GPT-5.1persistence baseline · Anchor: GPT-5.1
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$1.25Not calibrated
OpenAI
GPT-5.2persistence baseline · Anchor: GPT-5.2
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$1.75Not calibrated
OpenAI
GPT-5.6 Terrapersistence baseline · Anchor: GPT-5.6 Terra
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$2.00Not calibrated
OpenAI
GPT-6 Solpersistence baseline · Anchor: GPT-6 Sol
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$2.00Not calibrated
OpenAI
o3persistence baseline · Anchor: o3
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$2.00Not calibrated
OpenAI
GPT-4opersistence baseline · Anchor: GPT-4o
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$2.50Not calibrated
OpenAI
GPT-5.4persistence baseline · Anchor: GPT-5.4
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$2.50Not calibrated
OpenAI
GPT-5.6 Solpersistence baseline · Anchor: GPT-5.6 Sol
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$4.00Not calibrated
OpenAI
GPT-5.5persistence baseline · Anchor: GPT-5.5
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$5.00Not calibrated
OpenAI
GPT-6 Astrapersistence baseline · Anchor: GPT-6 Astra
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$10.00Not calibrated
Anthropic
Claude Opus linehistorical ensemble · Anchor: Claude Opus 5.5
5 / 3Jul 2023 – Nov 9, 2026persistence: 0.441damped-trend: 0.530robust-trend: 0.552$3.67$2.71 – $4.00
Anthropic
Claude Sonnet linehistorical ensemble · Anchor: Claude Sonnet 5.5
4 / 2Mar 4, 2024 – Nov 9, 2026persistence: 0.203damped-trend: 0.203robust-trend: 0.203$1.96$1.69 – $2.00
Anthropic
Claude Haiku linehistorical ensemble · Anchor: Claude Haiku 5.5
3 / 1Jul 2023 – Nov 9, 2026persistence: 0.916damped-trend: 2.414robust-trend: 2.414$0.0954$0.0399 – $0.10
Anthropic
Claude Haiku 4.5persistence baseline · Anchor: Claude Haiku 4.5
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$1.00Not calibrated
Anthropic
Claude Sonnet 4.6persistence baseline · Anchor: Claude Sonnet 4.6
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$3.00Not calibrated
Anthropic
Claude Opus 4.6persistence baseline · Anchor: Claude Opus 4.6
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$5.00Not calibrated
Anthropic
Claude Opus 4.7persistence baseline · Anchor: Claude Opus 4.7
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$5.00Not calibrated
Anthropic
Claude Opus 4.8persistence baseline · Anchor: Claude Opus 4.8
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$5.00Not calibrated
Anthropic
Claude Opus 5persistence baseline · Anchor: Claude Opus 5
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$5.00Not calibrated
Anthropic
Claude Fable 5.1persistence baseline · Anchor: Claude Fable 5.1
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$10.00Not calibrated
Google
Gemini Flash linehistorical ensemble · Anchor: Gemini 2.5 Flash-Lite
3 / 1Aug 12, 2024 – Nov 9, 2026persistence: 0.981damped-trend: 5.492robust-trend: 0.981$0.10$0.10 – $0.14
Google
Gemini 3 Flash Previewpersistence baseline · Anchor: Gemini 3 Flash Preview
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$0.50Not calibrated
Google
Gemini 3.6 Flashpersistence baseline · Anchor: Gemini 3.6 Flash
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$0.75Not calibrated
Google
Gemini 3.8 Flashpersistence baseline · Anchor: Gemini 3.8 Flash
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$0.75Not calibrated
Google
Gemini 2.5 Propersistence baseline · Anchor: Gemini 2.5 Pro
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$1.25Not calibrated
Google
Gemini 3.1 Pro Previewpersistence baseline · Anchor: Gemini 3.1 Pro Preview
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$2.00Not calibrated
Mistral
Mistral Small linehistorical ensemble · Anchor: Mistral Small 4
3 / 1Sep 17, 2024 – Nov 9, 2026persistence: 0.405damped-trend: 1.161robust-trend: 1.161$0.15$0.13 – $0.15
Mistral
Codestralpersistence baseline · Anchor: Codestral
2 / 0Sep 17, 2024 – Nov 9, 2026persistence: not measured$0.30Not calibrated
Mistral
Mistral Large linepersistence baseline · Anchor: Mistral Large 4
2 / 0Sep 17, 2024 – Nov 9, 2026persistence: not measured$0.70Not calibrated
Mistral
Ministral 3 3Bpersistence baseline · Anchor: Ministral 3 3B
2 / 0Oct 16, 2024 – Nov 9, 2026persistence: not measured$0.10Not calibrated
Mistral
Ministral 3 8Bpersistence baseline · Anchor: Ministral 3 8B
2 / 0Oct 16, 2024 – Nov 9, 2026persistence: not measured$0.15Not calibrated
Mistral
Ministral 3 14Bpersistence baseline · Anchor: Ministral 3 14B
1 / 0Nov 9, 2026 – Nov 9, 2026persistence: not measured$0.20Not calibrated
DeepSeek
DeepSeek general / Pro linehistorical ensemble · Anchor: DeepSeek V4 Pro 0813
6 / 4Feb 8, 2025 – Nov 9, 2026persistence: 0.699damped-trend: 0.831robust-trend: 0.819$1.92$1.32 – $4.43
DeepSeek
DeepSeek Flash linepersistence baseline · Anchor: DeepSeek V4.1 Flash 1017
4 / 2Apr 24, 2026 – Nov 9, 2026persistence: 0.191$0.30Not calibrated
xAI
Grok general-model linehistorical ensemble · Anchor: Grok 4.7
4 / 2Dec 12, 2024 – Nov 9, 2026persistence: 0.000damped-trend: 0.000robust-trend: 0.000$2.00$2.00 – $2.00
Cohere
Command R efficiency linehistorical ensemble · Anchor: Command R7B 12-2024
3 / 1Aug 30, 2024 – Nov 9, 2026persistence: 0.000damped-trend: 0.717robust-trend: 0.717$0.0368$0.018 – $0.0375
Cohere
Command R+ linehistorical ensemble · Anchor: Command R+ 08-2024
3 / 1Aug 30, 2024 – Nov 9, 2026persistence: 0.000damped-trend: 0.000robust-trend: 0.000$2.50$2.50 – $2.50
Amazon
Nova Micro · AWS us-east-1historical ensemble · Anchor: Nova Micro
5 / 3Jan 9, 2025 – Nov 9, 2026persistence: 0.000damped-trend: 0.000robust-trend: 0.000$0.035$0.035 – $0.035
Amazon
Nova Lite · AWS us-east-1historical ensemble · Anchor: Nova Lite
5 / 3Jan 9, 2025 – Nov 9, 2026persistence: 0.000damped-trend: 0.000robust-trend: 0.000$0.06$0.06 – $0.06
Amazon
Nova Pro · AWS us-east-1historical ensemble · Anchor: Nova Pro
5 / 3Jan 9, 2025 – Nov 9, 2026persistence: 0.000damped-trend: 0.000robust-trend: 0.000$0.80$0.80 – $0.80
Alibaba
Qwen Plus · Singapore / Internationalpersistence baseline · Anchor: Qwen3.7-Plus
2 / 0Jul 14, 2025 – Nov 9, 2026persistence: not measured$0.40Not calibrated
Sourced historical quotes · USD per 1M tokens
ObservationModelInputOutputSource
Aug 22, 2022QuoteOpenAI
GPT-3 Ada
$0.80$0.80Original source ↗
Aug 22, 2022QuoteOpenAI
GPT-3 Babbage
$1.20$1.20Original source ↗
Aug 22, 2022QuoteOpenAI
GPT-3 Curie
$6.00$6.00Original source ↗
Aug 22, 2022QuoteOpenAI
GPT-3 DavinciPre-cut rate quoted in the 22 Aug announcement. One rate for prompt and completion; not a 2020 launch price.
$60.00$60.00Original source ↗
Sep 1, 2022Price changeOpenAI
GPT-3 Ada
$0.40$0.40Original source ↗
Sep 1, 2022Price changeOpenAI
GPT-3 Babbage
$0.50$0.50Original source ↗
Sep 1, 2022Price changeOpenAI
GPT-3 Curie
$2.00$2.00Original source ↗
Sep 1, 2022Price changeOpenAI
GPT-3 Davinci
$20.00$20.00Original source ↗
Mar 1, 2023ReleaseOpenAI
GPT-3.5 Turbo
$2.00$2.00Original source ↗
Jul 2023QuoteAnthropic
Claude 2July 2023 price sheet. Plotted at month start; not an exact-day price observation.
$11.02$32.68Original source ↗
Jul 2023QuoteAnthropic
Claude Instant
$1.63$5.51Original source ↗
Nov 6, 2023ReleaseOpenAI
GPT-3.5 Turbo 1106
$1.00$2.00Original source ↗
Nov 6, 2023QuoteOpenAI
GPT-4Derived from the dated vendor comparison: Turbo was 3× cheaper on input at $10/M and 2× cheaper on output at $30/M. This is a November observation, not a launch-date quote.
$30.00$60.00Original source ↗
Nov 6, 2023ReleaseOpenAI
GPT-4 Turbo
$10.00$30.00Original source ↗
Mar 4, 2024ReleaseAnthropic
Claude 3 Opus
$15.00$75.00Original source ↗
Mar 4, 2024ReleaseAnthropic
Claude 3 Sonnet
$3.00$15.00Original source ↗
Mar 13, 2024ReleaseAnthropic
Claude 3 Haiku
$0.25$1.25Original source ↗
Jun 21, 2024ReleaseAnthropic
Claude 3.5 SonnetDate displayed on the provider announcement.
$3.00$15.00Original source ↗
Jul 18, 2024ReleaseOpenAI
GPT-4o mini
$0.15$0.60Original source ↗
Aug 12, 2024Price changeGoogle
Gemini 1.5 FlashPrompts under 128K tokens.
$0.075$0.30Original source ↗
Aug 30, 2024ReleaseCohere
Command R+ 08-2024The R+ line is separate from Command R and R7B.
$2.50$10.00Original source ↗
Aug 30, 2024ReleaseCohere
Command R 08-2024Command R efficiency line. Versions and parameter sizes differ; this is not a constant-capability index.
$0.15$0.60Original source ↗
Sep 17, 2024Price changeMistral
Codestral
$0.20$0.60Original source ↗
Sep 17, 2024Price changeMistral
Mistral Large 2
$2.00$6.00Original source ↗
Sep 17, 2024Price changeMistral
Mistral NeMo
$0.15$0.15Original source ↗
Sep 17, 2024Price changeMistral
Mistral Small 24.09
$0.20$0.60Original source ↗
Oct 14, 2024QuoteGoogle
Gemini 1.5 Flash-8BAnnounced 3 Oct; paid-tier billing began 14 Oct. Under 128K context. This historical model is retired.
$0.0375$0.15Original source ↗
Oct 16, 2024ReleaseMistral
Ministral 3BOriginal 2024 Ministral 3B. Different generation and capability from Ministral 3 3B (2512).
$0.04$0.04Original source ↗
Oct 16, 2024ReleaseMistral
Ministral 8BOriginal 2024 Ministral 8B. Kept within its size family; not a fixed-quality index.
$0.10$0.10Original source ↗
Dec 12, 2024Price changexAI
Grok-2 1212
$2.00$10.00Original source ↗
Jan 9, 2025QuoteAmazon
Nova LiteAWS archived in-region standard on-demand rate card, published 2025-01-09. Term effective 2024-12-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.06$0.24Original source ↗
Jan 9, 2025QuoteAmazon
Nova MicroAWS archived in-region standard on-demand rate card, published 2025-01-09. Term effective 2024-12-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.035$0.14Original source ↗
Jan 9, 2025QuoteAmazon
Nova ProAWS archived in-region standard on-demand rate card, published 2025-01-09. Term effective 2024-12-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.80$3.20Original source ↗
Jan 20, 2025ReleaseDeepSeek
DeepSeek R1Reasoning-model launch. Kept separate from the general-model and Flash families.
$0.55$2.19Original source ↗
Feb 8, 2025QuoteDeepSeek
DeepSeek V3Announced standard post-promotion rates; cache-miss input. V3 launched 26 Dec 2024.
$0.27$1.10Original source ↗
Apr 4, 2025QuoteAmazon
Nova LiteApproximate prices quoted by AWS.
$0.06$0.24Original source ↗
Apr 4, 2025QuoteAmazon
Nova MicroApproximate prices as quoted in the dated AWS article, not a launch-price assertion.
$0.035$0.14Original source ↗
Apr 14, 2025ReleaseOpenAI
GPT-4.1
$2.00$8.00Original source ↗
Apr 14, 2025ReleaseOpenAI
GPT-4.1 mini
$0.40$1.60Original source ↗
Apr 14, 2025ReleaseOpenAI
GPT-4.1 nano
$0.10$0.40Original source ↗
May 2, 2025QuoteAmazon
Nova LiteAWS archived in-region standard on-demand rate card, published 2025-05-02. Term effective 2025-04-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.06$0.24Original source ↗
May 2, 2025QuoteAmazon
Nova MicroAWS archived in-region standard on-demand rate card, published 2025-05-02. Term effective 2025-04-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.035$0.14Original source ↗
May 2, 2025QuoteAmazon
Nova ProAWS archived in-region standard on-demand rate card, published 2025-05-02. Term effective 2025-04-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.80$3.20Original source ↗
May 22, 2025ReleaseAnthropic
Claude Opus 4
$15.00$75.00Original source ↗
May 22, 2025ReleaseAnthropic
Claude Sonnet 4
$3.00$15.00Original source ↗
Jul 10, 2025QuoteMistral
Mistral Small 3.1Mistral explicitly quotes Small 3.1 in its 10 July Devstral announcement. This is the quote date, not an inferred March launch price.
$0.10$0.30Original source ↗
Jul 14, 2025QuoteAlibaba
Qwen-Plus (non-thinking)Standard non-thinking rates quoted in the dated Alibaba Cloud USD notice. Thinking-mode and mainland CNY offers are excluded. This does not establish an April launch price.
$0.40$1.20Original source ↗
Aug 7, 2025ReleaseOpenAI
GPT-5
$1.25$10.00Original source ↗
Aug 7, 2025ReleaseOpenAI
GPT-5 mini
$0.25$2.00Original source ↗
Aug 7, 2025ReleaseOpenAI
GPT-5 nano
$0.05$0.40Original source ↗
Sep 5, 2025Price changeDeepSeek
DeepSeek V3.1Price increase effective 16:00 UTC; standard cache-miss input, off-peak discounts ended.
$0.56$1.68Original source ↗
Sep 19, 2025ReleasexAI
Grok 4 FastUnder 128K input tier; reasoning and non-reasoning variants shared this token rate.
$0.20$0.50Original source ↗
Sep 29, 2025Price changeDeepSeek
DeepSeek V3.2-ExpNew standard cache-miss rates effective 10:00 UTC. Experimental successor; not a fixed-capability comparison.
$0.28$0.42Original source ↗
Nov 24, 2025ReleaseAnthropic
Claude Opus 4.5
$5.00$25.00Original source ↗
Dec 18, 2025QuoteCohere
Command R+ 08-2024Vendor documentation at immutable 18 Dec 2025 commit; the documented rate is unchanged.
$2.50$10.00Original source ↗
Dec 18, 2025QuoteCohere
Command R 08-2024Rate in the vendor documentation at immutable commit 83d07951c17c21725c349d80c75dc67ef065f4d6, committed 18 Dec 2025. Not backdated from model ID or frontmatter.
$0.15$0.60Original source ↗
Dec 18, 2025QuoteCohere
Command R7B 12-2024Smaller R-family option in the same immutable 18 Dec 2025 documentation snapshot. Model size/capability differs from R08.
$0.0375$0.15Original source ↗
Dec 23, 2025QuoteAmazon
Nova LiteAWS archived in-region standard on-demand rate card, published 2025-12-23. Term effective 2025-12-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.06$0.24Original source ↗
Dec 23, 2025QuoteAmazon
Nova MicroAWS archived in-region standard on-demand rate card, published 2025-12-23. Term effective 2025-12-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.035$0.14Original source ↗
Dec 23, 2025QuoteAmazon
Nova ProAWS archived in-region standard on-demand rate card, published 2025-12-23. Term effective 2025-12-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.80$3.20Original source ↗
Apr 24, 2026ReleaseDeepSeek
DeepSeek V4 Flash PreviewOriginal V4 Flash standard cache-miss launch rates. Flash history starts here; V3 flagship history is a separate line.
$0.14$0.28Original source ↗
Apr 24, 2026ReleaseDeepSeek
DeepSeek V4 Pro PreviewOriginal V4 preview standard cache-miss rates. Subsequent temporary discounts are not silently applied to earlier quotes.
$1.74$3.48Original source ↗
May 4, 2026QuoteAmazon
Nova LiteAWS archived in-region standard on-demand rate card, published 2026-05-04. Term effective 2026-04-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.06$0.24Original source ↗
May 4, 2026QuoteAmazon
Nova MicroAWS archived in-region standard on-demand rate card, published 2026-05-04. Term effective 2026-04-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.035$0.14Original source ↗
May 4, 2026QuoteAmazon
Nova ProAWS archived in-region standard on-demand rate card, published 2026-05-04. Term effective 2026-04-01. No batch, cross-region, cache, priority or custom-model rates. Date is a published rate-card observation, not a release or measured charge.
$0.80$3.20Original source ↗
Jul 16, 2026ReleasexAI
Grok 4.5Standard API rates stated in the dated launch article. Input pricing is unchanged from the December 2024 Grok-2 quote.
$2.00$6.00Original source ↗
Aug 16, 2026Price changeDeepSeek
DeepSeek V4 Flash 0731Peak-hour cache-miss rates effective 16:00 UTC. Off-peak input $0.22 / output $0.66 is excluded from this standard/peak comparison.
$0.44$1.32Original source ↗
Aug 16, 2026Price changeDeepSeek
DeepSeek V4 Pro 0813Peak-hour cache-miss rates effective 16:00 UTC. Off-peak rates were half of these; model and time-tier changes are explicit.
$1.32$3.96Original source ↗
Sep 10, 2026Price changeDeepSeek
DeepSeek V4.1 FlashPeak-hour cache-miss rates effective 04:00 UTC. Off-peak input $0.15 / output $0.60 is a different billing tier.
$0.30$1.20Original source ↗
Sep 21, 2026ReleasexAI
Grok 4.7Standard short-context starting rates in the dated launch article; the fast variant is a separate, more expensive offer.
$2.00$6.00Original source ↗
Nov 9, 2026QuoteMistral
Codestral
$0.30$0.90Original source ↗
Nov 9, 2026QuoteCohere
Command R 08-2024Current vendor model specification; distinct from the deprecated March 2024 Command R.
$0.15$0.60Original source ↗
Nov 9, 2026QuoteCohere
Command R7B 12-2024Current vendor model specification. R7B is a smaller, cheaper R-family option; no equal-capability claim.
$0.0375$0.15Original source ↗
Nov 9, 2026QuoteCohere
Command R 03-2024Legacy March 2024 price retained from Cohere FAQ. Vendor deprecation notice removed this model from current forecast anchors; newer R08 and R7B offers are separate observations.
$0.50$1.50Original source ↗
Nov 9, 2026QuoteCohere
Command R+ 04-2024Legacy April 2024 price retained from Cohere FAQ; deprecated model is excluded from current forecast anchors.
$3.00$15.00Original source ↗
Nov 9, 2026QuoteCohere
Command R+ 08-2024
$2.50$10.00Original source ↗
Nov 9, 2026QuoteDeepSeek
DeepSeek V4.1 Flash 1017Peak-hour, cache-miss rate. The provider also offers 50% lower off-peak rates.
$0.30$1.20Original source ↗
Nov 9, 2026QuoteDeepSeek
DeepSeek V4 Pro 0813Peak-hour, cache-miss rate; off-peak is half.
$1.32$3.96Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Fable 5.1
$10.00$50.00Original source ↗
Nov 9, 2026QuoteGoogle
Gemini 2.5 Flash-Lite
$0.10$0.40Original source ↗
Nov 9, 2026QuoteGoogle
Gemini 2.5 ProAt or below 200K input. Deprecated; provider documents automatic routing to a newer model from 1 Dec 2026.
$1.25$10.00Original source ↗
Nov 9, 2026QuoteGoogle
Gemini 3 Flash Preview
$0.50$3.00Original source ↗
Nov 9, 2026QuoteGoogle
Gemini 3.1 Pro PreviewPrompts at or below 200K tokens.
$2.00$12.00Original source ↗
Nov 9, 2026QuoteGoogle
Gemini 3.6 FlashIntroductory price through 31 Dec 2026; documented to double on 1 Jan 2027.
$0.75$3.75Original source ↗
Nov 9, 2026QuoteGoogle
Gemini 3.8 FlashIntroductory price through 31 Dec 2026. Published scheduled rates from 1 Jan 2027 are $1.50 input / $7.50 output.
$0.75$3.75Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-4o
$2.50$10.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.1
$1.25$10.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.2
$1.75$14.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.4 mini
$0.75$4.50Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.4 nano
$0.20$1.25Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.4
$2.50$15.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.5
$5.00$30.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.6 Luna
$0.20$1.20Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.6 SolPromotional rate currently documented through 30 Nov 2026.
$4.00$20.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-5.6 Terra
$2.00$12.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-6 Astra
$10.00$50.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-6 Luna
$0.10$0.50Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-6 Sol
$2.00$10.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-6.1 Sol
$2.00$10.00Original source ↗
Nov 9, 2026QuoteOpenAI
GPT-6.1 Terra
$1.00$5.00Original source ↗
Nov 9, 2026QuotexAI
Grok 4.7
$2.00$6.00Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Haiku 4.5
$1.00$5.00Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Haiku 5.5Prompts up to 100K tokens; higher rates above that threshold.
$0.10$0.50Original source ↗
Nov 9, 2026QuoteAmazon
Nova LiteAWS us-east-1 in-region standard on-demand text rate. USD per 1K tokens converted to USD per 1M. Batch, cross-region, cache, priority and custom-model prices are excluded.
$0.06$0.24Original source ↗
Nov 9, 2026QuoteAmazon
Nova ProAWS us-east-1 in-region standard on-demand text rate. USD per 1K tokens converted to USD per 1M. Batch, cross-region, cache, priority and custom-model prices are excluded.
$0.80$3.20Original source ↗
Nov 9, 2026QuoteAmazon
Nova MicroAWS us-east-1 in-region standard on-demand text rate. USD per 1K tokens converted to USD per 1M. Batch, cross-region, cache, priority and custom-model prices are excluded.
$0.035$0.14Original source ↗
Nov 9, 2026QuoteMistral
Ministral 3 14B
$0.20$0.20Original source ↗
Nov 9, 2026QuoteMistral
Ministral 3 3B
$0.10$0.10Original source ↗
Nov 9, 2026QuoteMistral
Ministral 3 8B
$0.15$0.15Original source ↗
Nov 9, 2026QuoteMistral
Mistral Small 4
$0.15$0.60Original source ↗
Nov 9, 2026QuoteMistral
Mistral Large 4
$0.70$2.10Original source ↗
Nov 9, 2026QuoteOpenAI
o3
$2.00$8.00Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Opus 4.6
$5.00$25.00Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Opus 4.7
$5.00$25.00Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Opus 4.8
$5.00$25.00Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Opus 5
$5.00$25.00Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Opus 5.5
$4.00$20.00Original source ↗
Nov 9, 2026QuoteAlibaba
Qwen3.7-PlusSingapore international deployment; at or below 256K input. Other regional and promotional rates differ.
$0.40$1.60Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Sonnet 4.6
$3.00$15.00Original source ↗
Nov 9, 2026QuoteAnthropic
Claude Sonnet 5.5
$2.00$10.00Original source ↗
01 / 07OBSERVATIONS · ASSUMPTIONS · SOURCES
06 / THE OPERATING MODEL

Study the incumbent.
Build for the gap.

Niche arbitrage: find a job people already pay for, understand where the existing product gets in the way, and build a focused alternative. That is the factory’s selection rule.

Start where someone already pays.

Study the incumbent, the pricing, and the job customers are trying to do. Look for a specific workflow with an identifiable buyer.

Build small. Charge for the job. Reinvest in what gets used.

The products.
Each with
its own job.

See what each one does.

INSIDE THE FACTORY / FOR INDIE HACKERS

Make the next bet
with the evidence.

I’ll be your test subject.

The models, the infrastructure, the acquisition tests. See what it took to build the products, what happened, and what still needs proving.

See the plans
Start with a real experiment. Follow the next one.