{
  "snapshots": ["2026-02", "2026-03", "2026-04", "2026-05", "2026-06", "2026-07", "2026-08", "2026-09", "2026-10", "2026-11"],
  "events": [
    { "date": "2026-03-01", "type": "price-increase", "description": "Haiku 4.5: input +25%, output +25%", "modelIds": ["haiku-4.5"], "prices": {"haiku-4.5": {"in": 1, "out": 5, "was": {"in": 0.8, "out": 4}}} },
    { "date": "2026-03-01", "type": "price-cut", "description": "Gemini 2.0 Flash: input -25%, output -25%", "modelIds": ["gemini-2.0-flash"] },
    { "date": "2026-03-01", "type": "price-cut", "description": "Mistral Large: input -17%, output -17%", "modelIds": ["mistral-large"], "prices": {"mistral-large": {"in": 0.5, "out": 1.5, "was": {"in": 0.6, "out": 1.8}}} },
    { "date": "2026-03-01", "type": "price-cut", "description": "Llama 3.3 70B: input -19%, output -19%", "modelIds": ["llama-3.3-70b"], "prices": {"llama-3.3-70b": {"in": 0.71, "out": 0.71, "was": {"in": 0.88, "out": 0.88}}} },
    { "date": "2026-04-01", "type": "price-increase", "description": "Gemini 2.5 Flash: output now includes thinking tokens ($0.60 → $2.50/MTok)", "modelIds": ["gemini-2.5-flash"], "prices": {"gemini-2.5-flash": {"in": 0.3, "out": 2.5, "was": {"in": 0.3, "out": 0.6}}} },
    { "date": "2026-04-01", "type": "correction", "description": "Gemini 2.0 Flash is $0.10/$0.40 per MTok, not $0.075/$0.15 as we listed (our listing was wrong).", "modelIds": ["gemini-2.0-flash"] },
    { "date": "2026-04-01", "type": "price-cut", "description": "DeepSeek: unified pricing — V3 & R1 both now $0.28/$0.42, R1 output -81%", "modelIds": ["deepseek-v3", "deepseek-r1"], "prices": {"deepseek-v3": {"in": 0.28, "out": 0.42}, "deepseek-r1": {"in": 0.28, "out": 0.42, "was": {"in": 0.55, "out": 2.19}}} },
    { "date": "2026-04-01", "type": "price-increase", "description": "Llama 3.3 70B: reverted to $0.88/$0.88 on Together.ai", "modelIds": ["llama-3.3-70b"], "prices": {"llama-3.3-70b": {"in": 0.88, "out": 0.88, "was": {"in": 0.71, "out": 0.71}}} },
    { "date": "2026-04-02", "type": "launch", "description": "Google releases Gemma 4 open-source family under Apache 2.0 (E2B/E4B/26B/31B)", "modelIds": [] },
    { "date": "2026-04-05", "type": "launch", "description": "Meta releases Llama 4 Scout (10M context) and finalises Maverick", "modelIds": ["llama4-scout", "llama4-maverick"] },
    { "date": "2026-04-16", "type": "launch", "description": "Anthropic releases Claude Opus 4.7 — same $5/$25 price but new tokenizer uses up to 35% more tokens", "modelIds": ["opus-4.7"], "prices": {"opus-4.7": {"in": 5, "out": 25}} },
    { "date": "2026-04-17", "type": "launch", "description": "xAI launches Grok 4.3 Beta on SuperGrok Heavy ($300/mo subscription, no API token pricing yet)", "modelIds": [] },
    { "date": "2026-04-24", "type": "launch", "description": "DeepSeek releases V4 Flash and V4 Pro with 1M context window", "modelIds": ["deepseek-v4-flash", "deepseek-v4-pro"], "names": {"deepseek-v4-flash": "DeepSeek V4 Flash"} },
    { "date": "2026-04-24", "type": "price-cut", "description": "DeepSeek V4 Flash launches at $0.14/$0.28 — 50% cheaper than V3/R1's unified pricing", "modelIds": ["deepseek-v4-flash"], "prices": {"deepseek-v4-flash": {"in": 0.14, "out": 0.28}}, "names": {"deepseek-v4-flash": "DeepSeek V4 Flash"} },
    { "date": "2026-04-26", "type": "price-cut", "description": "DeepSeek cache-hit input drops to $0.0028/MTok (one-tenth of launch rates)", "modelIds": ["deepseek-v4-flash", "deepseek-v4-pro"], "names": {"deepseek-v4-flash": "DeepSeek V4 Flash"} },
    { "date": "2026-04-28", "type": "deprecation", "description": "Gemini 2.0 Flash deprecated — full shutdown June 1, 2026", "modelIds": ["gemini-2.0-flash"] },
    { "date": "2026-04-28", "type": "launch", "description": "Gemini 2.5 Flash-Lite GA at $0.10/$0.40", "modelIds": ["gemini-2.5-flash-lite"], "prices": {"gemini-2.5-flash-lite": {"in": 0.1, "out": 0.4}} },
    { "date": "2026-04-28", "type": "launch", "description": "Gemini 3.1 Pro Preview and Gemini 3 Flash Preview now available", "modelIds": ["gemini-3.1-pro", "gemini-3-flash"] },
    { "date": "2026-04-23", "type": "launch", "description": "OpenAI launches GPT-5.5 ($5/$30, 1M context) — new flagship above GPT-5.4", "modelIds": ["gpt-5.5"], "prices": {"gpt-5.5": {"in": 5, "out": 30}} },
    { "date": "2026-05-07", "type": "launch", "description": "Google makes Gemini 3.1 Flash-Lite generally available at $0.25/$1.50", "modelIds": ["gemini-3.1-flash-lite"], "prices": {"gemini-3.1-flash-lite": {"in": 0.25, "out": 1.5}} },
    { "date": "2026-05-15", "type": "deprecation", "description": "xAI retires Grok 4.1 Fast, Grok 4 and Grok 3 — slugs now redirect to Grok 4.3", "modelIds": ["grok-4.1-fast", "grok-4"] },
    { "date": "2026-05-15", "type": "launch", "description": "Grok 4.3 gets API token pricing ($1.25/$2.50, 1M context); xAI ships the Grok Build 0.1 coding model", "modelIds": ["grok-4.3", "grok-build-0.1"], "prices": {"grok-4.3": {"in": 1.25, "out": 2.5}} },
    { "date": "2026-05-15", "type": "price-cut", "description": "Grok 4.20: input $2→$1.25, output $6→$2.50 (matches Grok 4.3); context revised 2M→1M", "modelIds": ["grok-4.20"], "prices": {"grok-4.20": {"in": 1.25, "out": 2.5, "was": {"in": 2, "out": 6}}} },
    { "date": "2026-05-19", "type": "launch", "description": "Google launches Gemini 3.5 Flash ($1.50/$9) at I/O 2026", "modelIds": ["gemini-3.5-flash"], "prices": {"gemini-3.5-flash": {"in": 1.5, "out": 9}} },
    { "date": "2026-05-25", "type": "price-cut", "description": "DeepSeek makes the V4 Pro 75% price cut permanent — $0.435/$0.87 is now the standard rate", "modelIds": ["deepseek-v4-pro"], "prices": {"deepseek-v4-pro": {"in": 0.435, "out": 0.87}} },
    { "date": "2026-05-28", "type": "launch", "description": "Anthropic releases Claude Opus 4.8 — same $5/$25 price; new tokenizer carried over from 4.7", "modelIds": ["opus-4.8"], "prices": {"opus-4.8": {"in": 5, "out": 25}} },
    { "date": "2026-05-29", "type": "price-cut", "description": "Mistral: Codestral $0.4/$2→$0.3/$0.9, Mistral Small 4 $0.15/$0.6→$0.1/$0.3; Large 3 context to 256K", "modelIds": ["codestral", "mistral-small", "mistral-large"], "prices": {"codestral": {"in": 0.3, "out": 0.9, "was": {"in": 0.4, "out": 2}}} },
    { "date": "2026-05-29", "type": "launch", "description": "Mistral adds Medium 3.5 ($1.50/$7.50) and Magistral Medium 1.2 reasoning model ($2/$5)", "modelIds": ["mistral-medium-3.5", "magistral-medium"], "prices": {"mistral-medium-3.5": {"in": 1.5, "out": 7.5}, "magistral-medium": {"in": 2, "out": 5}} },
    { "date": "2026-05-29", "type": "price-cut", "description": "Llama 4 Maverick hosted rates fall to ~$0.15/$0.60 on DeepInfra/OpenRouter", "modelIds": ["llama4-maverick"], "prices": {"llama4-maverick": {"in": 0.15, "out": 0.6, "was": {"in": 0.275, "out": 1.1}}} },
    { "date": "2026-05-29", "type": "launch", "description": "Azure adds GPT-5.4 ($2.50/$15, 272K) as its newest hosted GPT; GPT-5.2 moves to previous-gen", "modelIds": ["azure-gpt-5.4"], "prices": {"azure-gpt-5.4": {"in": 2.5, "out": 15}} },
    { "date": "2026-06-01", "type": "deprecation", "description": "Gemini 2.0 Flash shutdown — removed from the comparison table", "modelIds": ["gemini-2.0-flash"] },
    { "date": "2026-06-02", "type": "correction", "description": "o4-mini is $1.10/$4.40 per MTok, not $4/$16 as we listed (our listing was wrong).", "modelIds": ["o4-mini"] },
    { "date": "2026-06-09", "type": "launch", "description": "Anthropic launches Claude Fable 5 ($10/$50, 1M context) — public Mythos-class model, new top tier above Opus", "modelIds": ["fable-5"], "prices": {"fable-5": {"in": 10, "out": 50}} },
    { "date": "2026-06-15", "type": "deprecation", "description": "Anthropic retires legacy Claude Opus 4 and Sonnet 4 — recommended replacements are Opus 4.8 and Sonnet 4.6", "modelIds": [] },
    { "date": "2026-07-31", "type": "deprecation", "description": "Mistral schedules Magistral Medium 1.2 for retirement; reasoning folds into Mistral Medium 3.5 / Small 4 via reasoning_effort", "modelIds": ["magistral-medium"] },
    { "date": "2026-07-01", "type": "launch", "description": "Anthropic Claude Sonnet 5 now tracked at its effective $2/$10 rate, launched as introductory pricing and made permanent on Aug 11, 2026; adaptive thinking on by default", "modelIds": ["sonnet-5"], "prices": {"sonnet-5": {"in": 2, "out": 10}} },
    { "date": "2026-07-01", "type": "launch", "description": "OpenAI adds GPT-5.5 Pro ($30/$180, 1M context) — highest-precision flagship variant", "modelIds": ["gpt-5.5-pro"], "prices": {"gpt-5.5-pro": {"in": 30, "out": 180}} },
    { "date": "2026-07-01", "type": "launch", "description": "Azure adds GPT-5.5 ($5/$30, 1M context) to Foundry Global Standard", "modelIds": ["azure-gpt-5.5"], "prices": {"azure-gpt-5.5": {"in": 5, "out": 30}} },
    { "date": "2026-07-01", "type": "price-cut", "description": "Llama 3.3 70B: commodity host rates fall to ~$0.23/$0.40 (DeepInfra); $0.88 is now Together-only", "modelIds": ["llama-3.3-70b"], "prices": {"llama-3.3-70b": {"in": 0.23, "out": 0.4, "was": {"in": 0.88, "out": 0.88}}} },
    { "date": "2026-07-01", "type": "launch", "description": "Meta budget slot updated to Llama 3.1 8B — 128K context, ~$0.03/$0.06 on hosted providers (supersedes Llama 3 8B / 8K)", "modelIds": ["llama-3-8b"], "prices": {"llama-3-8b": {"in": 0.03, "out": 0.06}} },
    { "date": "2026-07-01", "type": "correction", "description": "Mistral Small 4 is $0.15/$0.60 per MTok with a 256K-token context window, not $0.10/$0.30 and 128K as we listed (our listing was wrong).", "modelIds": ["mistral-small"] },
    { "date": "2026-07-01", "type": "correction", "description": "Gemini 3.1 Pro and Gemini 2.5 Pro have a 1M-token context window (our listing was wrong).", "modelIds": ["gemini-3.1-pro", "gemini-2.5-pro"] },
    { "date": "2026-07-01", "type": "correction", "description": "GPT-5.4 mini, GPT-5.4 nano, Azure GPT-5.2 and Azure GPT-5-mini have a 400K-token context window (our listing was wrong).", "modelIds": ["gpt-5.4-mini", "gpt-5.4-nano", "azure-gpt-5.2", "azure-gpt-5-mini"] },
    { "date": "2026-07-01", "type": "correction", "description": "Azure GPT-5.4 has a 1M-token context window (our listing was wrong).", "modelIds": ["azure-gpt-5.4"] },
    { "date": "2026-07-01", "type": "correction", "description": "Magistral Medium 1.2 has a 128K-token context window (our listing was wrong).", "modelIds": ["magistral-medium"] },
    { "date": "2026-07-01", "type": "deprecation", "description": "OpenAI o4-mini marked legacy — superseded by GPT-5 mini; still served on the API", "modelIds": ["o4-mini"] },
    { "date": "2026-07-08", "type": "launch", "description": "xAI launches Grok 4.5 - new flagship at $2/$6 (500K context; $4/$12 above 200K prompt tokens). Cached input $0.30, no batch discount, not yet available in the EU.", "modelIds": ["grok-4.5"], "prices": {"grok-4.5": {"in": 2, "out": 6}} },
    { "date": "2026-07-09", "type": "launch", "description": "OpenAI GPT-5.6 family GA - Sol $5/$30, Terra $2.50/$15, Luna $1/$6 (1.05M context). New cache-write charge at 1.25x input; cached reads 10% of input.", "modelIds": ["gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"], "prices": {"gpt-5.6-sol": {"in": 5, "out": 30}, "gpt-5.6-terra": {"in": 2.5, "out": 15}, "gpt-5.6-luna": {"in": 1, "out": 6}} },
    { "date": "2026-07-24", "type": "deprecation", "description": "DeepSeek retires R1-0528 from its first-party API (deepseek-reasoner alias deprecates 2026-07-24); removed from the comparison table", "modelIds": ["deepseek-r1-0528"] },
    { "date": "2026-07-09", "type": "launch", "description": "Meta opens its first paid first-party API — Muse Spark 1.1 at $1.25/$4.25 (1M context), US developers only in public preview", "modelIds": ["muse-spark-1.1"], "prices": {"muse-spark-1.1": {"in": 1.25, "out": 4.25}}, "names": {"muse-spark-1.1": "Muse Spark 1.1"} },
    { "date": "2026-07-15", "type": "launch", "description": "Mistral adds Devstral 2, an open-weight agentic coding model at $0.40/$2.00 (256K context), alongside the Mistral Vibe CLI", "modelIds": ["devstral-2"], "prices": {"devstral-2": {"in": 0.4, "out": 2}} },
    { "date": "2026-07-21", "type": "launch", "description": "Google launches Gemini 3.6 Flash ($1.50/$7.50, 1M context) — same input as 3.5 Flash with output cut from $9 to $7.50 — and Gemini 3.5 Flash-Lite ($0.30/$2.50) GA", "modelIds": ["gemini-3.6-flash", "gemini-3.5-flash-lite"], "prices": {"gemini-3.6-flash": {"in": 1.5, "out": 7.5}, "gemini-3.5-flash-lite": {"in": 0.3, "out": 2.5}} },
    { "date": "2026-07-23", "type": "deprecation", "description": "OpenAI confirms o4-mini API shutdown for 2026-10-23; recommended migration is GPT-5.6 Terra. Still listed at $1.10/$4.40 on the standard pricing table until then", "modelIds": ["o4-mini", "azure-o4-mini"] },
    { "date": "2026-07-24", "type": "launch", "description": "Anthropic launches Claude Opus 5 ($5/$25, 1M context) — new Opus flagship at the same price as Opus 4.8, with fast mode at $10/$50 and a 512-token prompt-cache minimum", "modelIds": ["opus-5"], "prices": {"opus-5": {"in": 5, "out": 25}} },
    { "date": "2026-07-25", "type": "launch", "description": "Perplexity Sonar Deep Research now tracked at $2/$8, plus $2/MTok citation tokens, $3/MTok reasoning tokens and $5 per 1,000 searches", "modelIds": ["sonar-deep-research"], "prices": {"sonar-deep-research": {"in": 2, "out": 8}} },
    { "date": "2026-07-30", "type": "price-cut", "description": "OpenAI cuts GPT-5.6 Luna 80% ($1/$6 to $0.20/$1.20) and GPT-5.6 Terra 20% ($2.50/$15 to $2/$12), three weeks after the family's GA. Luna is now OpenAI's cheapest model and Terra undercuts GPT-5.4 on both input and output", "modelIds": ["gpt-5.6-luna", "gpt-5.6-terra"], "prices": {"gpt-5.6-luna": {"in": 0.2, "out": 1.2, "was": {"in": 1, "out": 6}}, "gpt-5.6-terra": {"in": 2, "out": 12, "was": {"in": 2.5, "out": 15}}} },
    { "date": "2026-07-31", "type": "deprecation", "description": "Mistral retires Magistral Medium 1.2, Magistral Small 1.2 and Devstral 2 on the same day. Magistral Medium left the comparison table then; Magistral Small (added in error on Aug 4) and Devstral 2 were removed from our tables in September 2026", "modelIds": ["magistral-medium", "magistral-small", "devstral-2"] },
    { "date": "2026-08-04", "type": "price-cut", "description": "Meta hosted rates re-baselined against DeepInfra: Llama 3.3 70B falls to $0.10/$0.32 (-57% input) and Llama 3.1 8B to $0.02/$0.04, the cheapest rate tracked", "modelIds": ["llama-3.3-70b", "llama-3-8b"], "prices": {"llama-3.3-70b": {"in": 0.1, "out": 0.32, "was": {"in": 0.23, "out": 0.4}}, "llama-3-8b": {"in": 0.02, "out": 0.04, "was": {"in": 0.03, "out": 0.06}}} },
    { "date": "2026-08-04", "type": "price-increase", "description": "Meta hosted rates rise for the Llama 4 pair: Maverick to $0.20/$0.80 (+33% input) and Scout to $0.10/$0.30 (+25% input)", "modelIds": ["llama4-maverick", "llama4-scout"], "prices": {"llama4-maverick": {"in": 0.2, "out": 0.8, "was": {"in": 0.15, "out": 0.6}}, "llama4-scout": {"in": 0.1, "out": 0.3, "was": {"in": 0.08, "out": 0.3}}} },
    { "date": "2026-08-04", "type": "tier-change", "description": "DeepSeek announces 2x peak-hour pricing for 09:00-12:00 and 14:00-18:00 Beijing time; start date not yet confirmed, so listed rates are unchanged", "modelIds": ["deepseek-v4-flash", "deepseek-v4-pro"], "names": {"deepseek-v4-flash": "DeepSeek V4 Flash"} },
    { "date": "2026-08-10", "type": "price-cut", "description": "Anthropic makes Claude Sonnet 5's $2/$10 launch rate permanent — the scheduled Sep 1 increase to $3/$15 will not occur. Sonnet 5 was previously tracked at its $3/$15 list price", "modelIds": ["sonnet-5"], "prices": {"sonnet-5": {"in": 2, "out": 10}} },
    { "date": "2026-08-12", "type": "launch", "description": "xAI launches Grok 4.6 as its new flagship at $2/$6 (500K context; $4/$12 above 200K prompt tokens), the same token rates as Grok 4.5 but with cached input at $0.50 instead of $0.30. No batch discount", "modelIds": ["grok-4.6", "grok-4.5"], "prices": {"grok-4.6": {"in": 2, "out": 6}} },
    { "date": "2026-08-13", "type": "price-cut", "description": "Google halves Gemini 3.6 Flash to an introductory $0.75/$3.75 through Dec 31, 2026 (the $1.50/$7.50 list price returns Jan 1, 2027) and ships Gemini 3.7 Flash GA at the same introductory rate", "modelIds": ["gemini-3.6-flash", "gemini-3.7-flash"], "prices": {"gemini-3.6-flash": {"in": 0.75, "out": 3.75, "was": {"in": 1.5, "out": 7.5}}, "gemini-3.7-flash": {"in": 0.75, "out": 3.75}} },
    { "date": "2026-08-16", "type": "price-increase", "description": "DeepSeek raises prices and starts peak/off-peak billing, with off-peak at half the peak rate (peak is 01:00-04:00 and 06:00-10:00 UTC on weekdays). Peak rates: V4 Flash $0.44/$1.32 (from $0.14/$0.28) and V4 Pro $1.32/$3.96 (from $0.435/$0.87). DeepSeek is now tracked at peak rates", "modelIds": ["deepseek-v4-flash", "deepseek-v4-pro"], "prices": {"deepseek-v4-flash": {"in": 0.44, "out": 1.32, "was": {"in": 0.14, "out": 0.28}}, "deepseek-v4-pro": {"in": 1.32, "out": 3.96, "was": {"in": 0.435, "out": 0.87}}}, "names": {"deepseek-v4-flash": "DeepSeek V4 Flash"} },
    { "date": "2026-08-21", "type": "price-cut", "description": "OpenAI puts GPT-5.6 Sol on promotional pricing of $4/$20 (from $5/$30; cached input $0.40), available at least through Nov 21, 2026", "modelIds": ["gpt-5.6-sol"], "prices": {"gpt-5.6-sol": {"in": 4, "out": 20, "was": {"in": 5, "out": 30}}} },
    { "date": "2026-09-01", "type": "launch", "description": "Anthropic launches Claude Fable 5.1 at the same $10/$50 as Fable 5, with cache reads cut 75% to $0.25/MTok; Fable 5 moves to legacy", "modelIds": ["fable-5.1", "fable-5"], "prices": {"fable-5.1": {"in": 10, "out": 50}} },
    { "date": "2026-09-02", "type": "launch", "description": "Google launches Gemini 3.8 Flash GA at an introductory $0.75/$3.75 (1M context), with the $1.50/$7.50 list price from Jan 1, 2027", "modelIds": ["gemini-3.8-flash"], "prices": {"gemini-3.8-flash": {"in": 0.75, "out": 3.75}} },
    { "date": "2026-09-02", "type": "launch", "description": "Meta releases Muse Spark 1.3 at an unchanged $1.25/$4.25; the tracked Muse Spark row moves from 1.1 to 1.3. Muse Spark 1.2 shipped Aug 5 at the same price and widened access beyond the US", "modelIds": ["muse-spark-1.1"], "prices": {"muse-spark-1.1": {"in": 1.25, "out": 4.25}} },
    { "date": "2026-09-03", "type": "launch", "description": "OpenAI launches GPT-6 Astra, its most capable model, at $10/$50 (1.05M context; $20/$75 above 272K; 50% batch discount). Azure Foundry lists it at the same Global Standard rate, without batch pricing", "modelIds": ["gpt-6-astra", "azure-gpt-6-astra"], "prices": {"gpt-6-astra": {"in": 10, "out": 50}, "azure-gpt-6-astra": {"in": 10, "out": 50}} },
    { "date": "2026-09-10", "type": "price-cut", "description": "DeepSeek V4.1 Flash replaces V4 Flash and cuts Flash rates to $0.30/$1.20 at peak ($0.15/$0.60 off-peak). V4 Pro's planned Sep 14 retirement is reversed", "modelIds": ["deepseek-v4-flash", "deepseek-v4-pro"], "prices": {"deepseek-v4-flash": {"in": 0.3, "out": 1.2, "was": {"in": 0.44, "out": 1.32}}} },
    { "date": "2026-09-15", "type": "deprecation", "description": "Perplexity will support its Sonar Chat Completions models only until Sep 27, 2026, as it moves users to its Agent API", "modelIds": ["sonar", "sonar-pro", "sonar-reasoning", "sonar-deep-research"] },
    { "date": "2026-09-15", "type": "correction", "description": "Codestral's context window is 128K, not 256K (our listing was wrong).", "modelIds": ["codestral"] },
    { "date": "2026-09-15", "type": "correction", "description": "GPT-5.5, GPT-5.5 Pro and GPT-5.4 have a 1.05M-token context window, not 1M as we listed.", "modelIds": ["gpt-5.5", "gpt-5.5-pro", "gpt-5.4"] },
    { "date": "2026-09-15", "type": "correction", "description": "Azure retires o4-mini on 19 Nov 2026, not 23 Oct as we listed.", "modelIds": ["azure-o4-mini"] },
    { "date": "2026-09-18", "type": "tier-change", "description": "Google limits the Gemini 2.5 models to developers who have used them before. They are not deprecated and keep their prices, but new projects are steered to Gemini 3.5 Flash-Lite and 3.8 Flash", "modelIds": ["gemini-2.5-pro", "gemini-2.5-flash", "gemini-2.5-flash-lite"] },
    { "date": "2026-09-21", "type": "launch", "description": "xAI launches Grok 4.7 as its new flagship at $2/$6 (500K context; $4/$12 above 200K prompt tokens), the same rates as Grok 4.6 including $0.50 cached input. No batch discount", "modelIds": ["grok-4.7", "grok-4.6"], "prices": {"grok-4.7": {"in": 2, "out": 6}} },
    { "date": "2026-09-22", "type": "launch", "description": "Anthropic launches Claude Opus 5.5 at $4/$20 (1M context), 20% below Opus 5 on both input and output, with cache reads at $0.20/MTok (0.05x input) and fast mode at $8/$40. Opus 5 stays available at $5/$25", "modelIds": ["opus-5.5", "opus-5"], "prices": {"opus-5.5": {"in": 4, "out": 20}} },
    { "date": "2026-09-22", "type": "launch", "description": "OpenAI launches GPT-6 Luna at $0.10/$0.50 (1.05M context), its cheapest current-generation model, alongside GPT-6 Sol at $2/$10", "modelIds": ["gpt-6-luna"], "prices": {"gpt-6-luna": {"in": 0.1, "out": 0.5}} },
    { "date": "2026-09-27", "type": "tier-change", "description": "Perplexity ends Sonar Chat Completions support. Sonar moves to the Agent API at $0.25/$2.50 (from $1/$1) with web search billed separately at $2.50 per 1,000 calls; Sonar Pro, Reasoning Pro and Deep Research become Agent API presets billed at the underlying model's rates, and are no longer tracked", "modelIds": ["sonar", "sonar-pro", "sonar-reasoning", "sonar-deep-research"] },
    { "date": "2026-09-28", "type": "launch", "description": "Anthropic launches Claude Sonnet 5.5 at $2/$10 (1M context), the same price as the Sonnet 5 it replaces", "modelIds": ["sonnet-5.5", "sonnet-5"], "prices": {"sonnet-5.5": {"in": 2, "out": 10}} },
    { "date": "2026-09-29", "type": "launch", "description": "OpenAI launches GPT-6.1 Sol at $2/$10 (1.05M context; cached input $0.10), replacing GPT-6 Sol a week after launch and matching Claude Sonnet 5.5's price exactly", "modelIds": ["gpt-6.1-sol"], "prices": {"gpt-6.1-sol": {"in": 2, "out": 10}} },
    { "date": "2026-10-01", "type": "deprecation", "description": "OpenAI deprecates GPT-5.4 Nano, with its API shutting down on Apr 1, 2027 and GPT-6 Luna as the recommended replacement", "modelIds": ["gpt-5.4-nano"] },
    { "date": "2026-10-01", "type": "price-increase", "description": "DeepInfra retires Llama 4 Maverick, so the tracked rate moves to Novita at $0.27/$0.85 (from $0.20/$0.80), keeping the full 1M context", "modelIds": ["llama4-maverick"], "prices": {"llama4-maverick": {"in": 0.27, "out": 0.85, "was": {"in": 0.2, "out": 0.8}}} },
    { "date": "2026-10-05", "type": "correction", "description": "Azure GPT-5.5 and Azure GPT-5.4 have a 1.05M-token context window, not 1M as we listed.", "modelIds": ["azure-gpt-5.5", "azure-gpt-5.4"] },
    { "date": "2026-10-05", "type": "tier-change", "description": "The Manus free plan refreshes 300 credits a day, capped at 1,500 a month, with no starter pack.", "modelIds": [] },
    { "date": "2026-10-05", "type": "tier-change", "description": "DeepSeek's off-peak rate now also covers Chinese public holidays in full.", "modelIds": ["deepseek-v4-flash", "deepseek-v4-pro"] },
    { "date": "2026-10-06", "type": "launch", "description": "Mistral launches Mistral Large 4 in public preview, a multimodal model with a 1M context window, at a $1.36/$4.18 list price after a two-week launch discount", "modelIds": ["mistral-large-4"], "prices": {"mistral-large-4": {"in": 1.36, "out": 4.18}} },
    { "date": "2026-10-07", "type": "launch", "description": "Anthropic launches Claude Haiku 5.5 at $0.10/$0.50 for prompts up to 100K tokens, a tenth of Haiku 4.5's $1/$5. A prompt over 100K tokens, cache reads and writes included, bills the whole request at $0.50/$2.50. 1M context, 128K max output, 50% batch discount. It ties GPT-6 Luna on short prompts; Haiku 4.5 moves to legacy", "modelIds": ["haiku-5.5", "haiku-4.5"], "prices": {"haiku-5.5": {"in": 0.1, "out": 0.5}} },
    { "date": "2026-10-07", "type": "price-cut", "description": "Anthropic halves Claude Sonnet 5.5 cache reads from $0.20 to $0.10/MTok (0.05x input, the same multiplier as Opus 5.5). Token rates and cache writes are unchanged; Sonnet 5.5 now matches GPT-6.1 Sol's $0.10 cached input", "modelIds": ["sonnet-5.5"] },
    { "date": "2026-10-08", "type": "deprecation", "description": "Google deprecates Gemini 3.5 Flash and routes its API requests to Gemini 3.6 Flash, its replacement; removed from our tables", "modelIds": ["gemini-3.5-flash", "gemini-3.6-flash"] },
    { "date": "2026-10-13", "type": "launch", "description": "Google launches Gemini 4 Argon GA, its most capable model, at an introductory $2/$10 (1M context) through Nov 30, 2026, with the $4/$20 list price from Dec 1, 2026", "modelIds": ["gemini-4-argon"], "prices": {"gemini-4-argon": {"in": 2, "out": 10}} },
    { "date": "2026-10-23", "type": "deprecation", "description": "OpenAI shuts down o4-mini; removed from our tables. GPT-5.6 Terra is OpenAI's recommended replacement. Azure keeps serving its o4-mini until 19 Nov 2026", "modelIds": ["o4-mini"] },
    { "date": "2026-10-26", "type": "launch", "description": "OpenAI launches GPT-6.1 Terra at $1/$5 (1.05M context; cached input $0.05; $2/$7.50 above 272K)", "modelIds": ["gpt-6.1-terra"], "prices": {"gpt-6.1-terra": {"in": 1, "out": 5}} },
    { "date": "2026-10-26", "type": "correction", "description": "GPT-5.5 Pro bills $60/$270 for prompts over 272K input tokens (our listing was wrong).", "modelIds": ["gpt-5.5-pro"] },
    { "date": "2026-10-26", "type": "tier-change", "description": "xAI now lists Grok 4.7, Grok 4.6 and Grok 4.5 for its 20% batch discount", "modelIds": ["grok-4.7", "grok-4.6", "grok-4.5"] },
    { "date": "2026-10-26", "type": "tier-change", "description": "Meta now caps Muse Spark Contributor mode at 150 requests a minute, up from 100", "modelIds": ["muse-spark-1.1"] },
    { "date": "2026-10-30", "type": "price-cut", "description": "Mistral now lists Mistral Large 4 at $0.70/$2.10 (from $1.36/$4.18), with cached input at $0.07", "modelIds": ["mistral-large-4"], "prices": {"mistral-large-4": {"in": 0.7, "out": 2.1, "was": {"in": 1.36, "out": 4.18}}} },
    { "date": "2026-10-31", "type": "deprecation", "description": "Mistral deprecates Mistral Large 3 and Mistral Medium 3.5, with both retiring from its API on Apr 30, 2027 and Mistral Large 4 as the named alternative", "modelIds": ["mistral-large", "mistral-medium-3.5", "mistral-large-4"] },
    { "date": "2026-10-31", "type": "correction", "description": "Gemini 3.1 Flash-Lite became generally available on May 7, 2026, not May 19 (our listing was wrong).", "modelIds": ["gemini-3.1-flash-lite"] },
    { "date": "2026-11-02", "type": "deprecation", "description": "Google deprecates Gemini 2.5 Pro and Gemini 2.5 Flash, with a Dec 1, 2026 shutdown date: from then their API requests route to Gemini 3.8 Flash and Gemini 3.5 Flash-Lite respectively, billed at those models' rates", "modelIds": ["gemini-2.5-pro", "gemini-2.5-flash", "gemini-3.8-flash", "gemini-3.5-flash-lite"] },
    { "date": "2026-11-05", "type": "tier-change", "description": "OpenAI now says GPT-5.6 Sol's promotional pricing is available until Nov 30, 2026, having said earlier that it runs at least through Nov 21", "modelIds": ["gpt-5.6-sol"] }
  ]
}