33 events in a year. Then 155 in the next.
Every model-market event Vitna has recorded since Mar 2023, newest first. Each entry links to a primary source on a domain the provider itself publishes on, and carries the date that source was last fetched; the newest of those checks is 2026-11-19. The log only ever gains entries.
Every event at its true date, newest left.
Each move as a share of the price before it.
83 detected moves carry both of their periods, so each bar is the move as a share of the price before it: cuts run left, rises run right, and the longest bar is the largest move. Each model leads to its own page and each meter to the move's own page. Curated entries cite an announcement rather than two readings and list below without a bar.
Every entry, with the document behind it.
216 events, Mar 2023 to Nov 2026.
- 2026Novita AIprice · detectedsource ↗
Novita AI Kimi K3 cache write price, 1 hour retention falls to $0.30
Cache write price, 1 hour retention for Novita AI Kimi K3 moves from $6.00 to $0.30 in USD per 1M tokens, a 95 percent cut. Read on 19 Nov 2026 at 19 Nov 09:08 UTC; the documents before and after are archived as 8a4013d3a0d6 and fdf7389b43c2.
- Novita AIprice · detectedsource ↗
Novita AI Kimi K3 cache write price, 5 minute retention rises to $6.00
Cache write price, 5 minute retention for Novita AI Kimi K3 moves from $3.00 to $6.00 in USD per 1M tokens, a 100 percent rise. Read on 19 Nov 2026 at 19 Nov 09:08 UTC; the documents before and after are archived as 8a4013d3a0d6 and fdf7389b43c2.
- Novita AIprice · detectedsource ↗
Novita AI Kimi K3 cached input price rises to $3.00
Cached input price for Novita AI Kimi K3 moves from $0.30 to $3.00 in USD per 1M tokens, a 900 percent rise. Read on 19 Nov 2026 at 19 Nov 09:08 UTC; the documents before and after are archived as 8a4013d3a0d6 and fdf7389b43c2.
- Novita AIprice · detectedsource ↗
Novita AI Qwen3.8-Max cache write price, 5 minute retention falls to $0.25
Cache write price, 5 minute retention for Novita AI Qwen3.8-Max moves from $2.50 to $0.25 in USD per 1M tokens, a 90 percent cut. Read on 19 Nov 2026 at 19 Nov 09:08 UTC; the documents before and after are archived as 8a4013d3a0d6 and fdf7389b43c2.
- Novita AIprice · detectedsource ↗
Novita AI Qwen3.8-Max cached input price rises to $2.50
Cached input price for Novita AI Qwen3.8-Max moves from $0.25 to $2.50 in USD per 1M tokens, a 900 percent rise. Read on 19 Nov 2026 at 19 Nov 09:08 UTC; the documents before and after are archived as 8a4013d3a0d6 and fdf7389b43c2.
- Novita AIprice · detectedsource ↗
Novita AI Qwen3.8-Flash cache write price, 5 minute retention falls to $0.016
Cache write price, 5 minute retention for Novita AI Qwen3.8-Flash moves from $0.20 to $0.016 in USD per 1M tokens, a 92 percent cut. Read on 19 Nov 2026 at 19 Nov 09:08 UTC; the documents before and after are archived as 8a4013d3a0d6 and fdf7389b43c2.
- Novita AIprice · detectedsource ↗
Novita AI Qwen3.8-Flash cached input price rises to $0.20
Cached input price for Novita AI Qwen3.8-Flash moves from $0.016 to $0.20 in USD per 1M tokens, a 1150 percent rise. Read on 19 Nov 2026 at 19 Nov 09:08 UTC; the documents before and after are archived as 8a4013d3a0d6 and fdf7389b43c2.
- Together AIprice · detectedsource ↗
Together AI Qwen Image 2.0 per image falls to $0.035
Per image for Together AI Qwen Image 2.0 moves from $0.04 to $0.035 in USD per image, a 12 percent cut. Read on 16 Nov 2026 at 16 Nov 05:03 UTC; the documents before and after are archived as 482476d1d32a and cfe43b240987.
- Together AIprice · detectedsource ↗
Together AI Qwen Image 2.0 Pro per image falls to $0.075
Per image for Together AI Qwen Image 2.0 Pro moves from $0.08 to $0.075 in USD per image, a 6 percent cut. Read on 16 Nov 2026 at 16 Nov 05:03 UTC; the documents before and after are archived as 482476d1d32a and cfe43b240987.
- Together AIprice · detectedsource ↗
Together AI Gemini 3.1 Flash Image (Nano Banana 2) per image falls to $0.0466
Per image for Together AI Gemini 3.1 Flash Image (Nano Banana 2) moves from $0.05 to $0.0466 in USD per image, a 7 percent cut. Read on 16 Nov 2026 at 16 Nov 05:03 UTC; the documents before and after are archived as 482476d1d32a and cfe43b240987.
- Deepgramprice · detectedsource ↗
Deepgram Nova-3 Multilingual streaming transcription falls to $0.0058
Streaming transcription for Deepgram Nova-3 Multilingual moves from $0.0092 to $0.0058 in USD per audio-minute, a 37 percent cut. Read on 16 Nov 2026 at 16 Nov 05:03 UTC; the documents before and after are archived as 16fde26b52e0 and df69068bdb5e.
- Deepgramprice · detectedsource ↗
Deepgram Nova-3 Monolingual streaming transcription falls to $0.0048
Streaming transcription for Deepgram Nova-3 Monolingual moves from $0.0077 to $0.0048 in USD per audio-minute, a 38 percent cut. Read on 16 Nov 2026 at 16 Nov 05:03 UTC; the documents before and after are archived as 16fde26b52e0 and df69068bdb5e.
- Deepgramprice · detectedsource ↗
Deepgram Flux English streaming transcription falls to $0.0065
Streaming transcription for Deepgram Flux English moves from $0.0077 to $0.0065 in USD per audio-minute, a 16 percent cut. Read on 16 Nov 2026 at 16 Nov 05:03 UTC; the documents before and after are archived as 16fde26b52e0 and df69068bdb5e.
- Lambdaprice · detectedsource ↗
Lambda A100 80GB on-demand list price rises to $3.19
On-demand list price for Lambda A100 80GB moves from $2.79 to $3.19 in USD per GPU-hour, a 14 percent rise. Read on 16 Nov 2026 at 16 Nov 00:02 UTC; the documents before and after are archived as eb0161ea81bf and 3a277b420f1d.
- Lambdaprice · detectedsource ↗
Lambda B200 on-demand list price rises to $7.69
On-demand list price for Lambda B200 moves from $6.69 to $7.69 in USD per GPU-hour, a 15 percent rise. Read on 16 Nov 2026 at 16 Nov 00:02 UTC; the documents before and after are archived as eb0161ea81bf and 3a277b420f1d.
- Lambdaprice · detectedsource ↗
Lambda H100 SXM on-demand list price rises to $4.59
On-demand list price for Lambda H100 SXM moves from $3.99 to $4.59 in USD per GPU-hour, a 15 percent rise. Read on 16 Nov 2026 at 16 Nov 00:02 UTC; the documents before and after are archived as eb0161ea81bf and 3a277b420f1d.
- RunPodprice · detectedsource ↗
RunPod H100 SXM on-demand list price rises to $4.39
On-demand list price for RunPod H100 SXM moves from $3.99 to $4.39 in USD per GPU-hour, a 10 percent rise. Read on 13 Nov 2026 at 13 Nov 01:05 UTC; the documents before and after are archived as c53aea925c60 and e39fe7fa3fae.
- LlamaIndexprice · detectedsource ↗
LlamaIndex LlamaParse Agentic Plus structure from a page falls to $0.0375
Structure from a page for LlamaIndex LlamaParse Agentic Plus moves from $0.0563 to $0.0375 in USD per page, a 33 percent cut. Read on 7 Nov 2026 at 7 Nov 21:07 UTC; the documents before and after are archived as 7696bcfa166a and 48aed3d7ea06.
- LlamaIndexprice · detectedsource ↗
LlamaIndex LlamaParse Agentic structure from a page rises to $0.015
Structure from a page for LlamaIndex LlamaParse Agentic moves from $0.0125 to $0.015 in USD per page, a 20 percent rise. Read on 7 Nov 2026 at 7 Nov 21:07 UTC; the documents before and after are archived as 7696bcfa166a and 48aed3d7ea06.
- LlamaIndexprice · detectedsource ↗
LlamaIndex LlamaParse Cost-effective structure from a page rises to $0.0045
Structure from a page for LlamaIndex LlamaParse Cost-effective moves from $0.00375 to $0.0045 in USD per page, a 20 percent rise. Read on 7 Nov 2026 at 7 Nov 21:07 UTC; the documents before and after are archived as 7696bcfa166a and 48aed3d7ea06.
- LlamaIndexprice · detectedsource ↗
LlamaIndex LlamaParse Fast text from a page rises to $0.0015
Text from a page for LlamaIndex LlamaParse Fast moves from $0.00125 to $0.0015 in USD per page, a 20 percent rise. Read on 7 Nov 2026 at 7 Nov 21:07 UTC; the documents before and after are archived as 7696bcfa166a and 48aed3d7ea06.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max output price rises to $7.50
Output price for Together AI Qwen3.7-Max moves from $4.50 to $7.50 in USD per 1M tokens, a 67 percent rise. Read on 6 Nov 2026 at 6 Nov 21:01 UTC; the documents before and after are archived as b47eaf4eaadd and 02f746fcebb7.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max cached input price rises to $0.50
Cached input price for Together AI Qwen3.7-Max moves from $0.30 to $0.50 in USD per 1M tokens, a 67 percent rise. Read on 6 Nov 2026 at 6 Nov 21:01 UTC; the documents before and after are archived as b47eaf4eaadd and 02f746fcebb7.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max input price rises to $2.50
Input price for Together AI Qwen3.7-Max moves from $1.50 to $2.50 in USD per 1M tokens, a 67 percent rise. Read on 6 Nov 2026 at 6 Nov 21:01 UTC; the documents before and after are archived as b47eaf4eaadd and 02f746fcebb7.
- DeepInfraprice · detectedsource ↗
DeepInfra NVIDIA-Nemotron-3-Super-120B-A12B input price falls to $0.075
Input price for DeepInfra NVIDIA-Nemotron-3-Super-120B-A12B moves from $0.085 to $0.075 in USD per 1M tokens, a 12 percent cut. Read on 3 Nov 2026 at 3 Nov 21:08 UTC; the documents before and after are archived as 3fdcbf08146d and 3224cd3e6bbb.
- Deepgramprice · detectedsource ↗
Deepgram Nova-3 Multilingual streaming transcription rises to $0.0092
Streaming transcription for Deepgram Nova-3 Multilingual moves from $0.0058 to $0.0092 in USD per audio-minute, a 59 percent rise. Read on 2 Nov 2026 at 2 Nov 22:09 UTC; the documents before and after are archived as a856b89b8123 and 0399a85e36ec.
- Deepgramprice · detectedsource ↗
Deepgram Nova-3 Monolingual streaming transcription rises to $0.0077
Streaming transcription for Deepgram Nova-3 Monolingual moves from $0.0048 to $0.0077 in USD per audio-minute, a 60 percent rise. Read on 2 Nov 2026 at 2 Nov 22:09 UTC; the documents before and after are archived as a856b89b8123 and 0399a85e36ec.
- Deepgramprice · detectedsource ↗
Deepgram Flux English streaming transcription rises to $0.0077
Streaming transcription for Deepgram Flux English moves from $0.0065 to $0.0077 in USD per audio-minute, a 18 percent rise. Read on 2 Nov 2026 at 2 Nov 22:09 UTC; the documents before and after are archived as a856b89b8123 and 0399a85e36ec.
- Speechmaticsprice · detectedsource ↗
Speechmatics Real-time Enhanced streaming transcription rises to $0.01
Corrected. The row that supersedes this one stands in the record.
Streaming transcription for Speechmatics Real-time Enhanced moves from $0.01 to $0.01 in USD per audio-minute, a 86 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 48d3d8603d1a and 753ffc82a8e6.
- Speechmaticspricesource ↗
Speechmatics Real-time Enhanced streaming transcription price is unchanged
A correction, superseding the row it replaces.
Speechmatics lists Real-time Enhanced streaming transcription at $0.80 an hour ($0.013333 a minute) on pay as you go, which its pricing page calls the listed hourly rate with no discounts. The page's own data gave the same rate on 10 September, when the record first read it, so no price moved on 26 October. Until then the record held $0.43 an hour, the figure the page's table printed after an opt-in 33 percent model-training discount and a 20 percent volume discount that applies only above 500 hours a month; on 26 October it read the undiscounted rate for the first time.
- Speechmaticsprice · detectedsource ↗
Speechmatics Real-time Standard streaming transcription rises to $0.01
Corrected. The row that supersedes this one stands in the record.
Streaming transcription for Speechmatics Real-time Standard moves from $0.00 to $0.01 in USD per audio-minute, a 87 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 48d3d8603d1a and 753ffc82a8e6.
- Speechmaticspricesource ↗
Speechmatics Real-time Standard streaming transcription price is unchanged
A correction, superseding the row it replaces.
Speechmatics lists Real-time Standard streaming transcription at $0.45 an hour ($0.0075 a minute) on pay as you go, which its pricing page calls the listed hourly rate with no discounts. The page's own data gave the same rate on 10 September, when the record first read it, so no price moved on 26 October. Until then the record held $0.24 an hour, the figure the page's table printed after an opt-in 33 percent model-training discount and a 20 percent volume discount that applies only above 500 hours a month; on 26 October it read the undiscounted rate for the first time.
- Speechmaticsprice · detectedsource ↗
Speechmatics Batch Enhanced pre-recorded transcription rises to $0.01
Corrected. The row that supersedes this one stands in the record.
Pre-recorded transcription for Speechmatics Batch Enhanced moves from $0.01 to $0.01 in USD per audio-minute, a 87 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 48d3d8603d1a and 753ffc82a8e6.
- Speechmaticspricesource ↗
Speechmatics Batch Enhanced pre-recorded transcription price is unchanged
A correction, superseding the row it replaces.
Speechmatics lists Batch Enhanced pre-recorded transcription at $0.75 an hour ($0.0125 a minute) on pay as you go, which its pricing page calls the listed hourly rate with no discounts. The page's own data gave the same rate on 10 September, when the record first read it, so no price moved on 26 October. Until then the record held $0.40 an hour, the figure the page's table printed after an opt-in 33 percent model-training discount and a 20 percent volume discount that applies only above 500 hours a month; on 26 October it read the undiscounted rate for the first time.
- Speechmaticsprice · detectedsource ↗
Speechmatics Batch Standard pre-recorded transcription rises to $0.01
Corrected. The row that supersedes this one stands in the record.
Pre-recorded transcription for Speechmatics Batch Standard moves from $0.00 to $0.01 in USD per audio-minute, a 87 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 48d3d8603d1a and 753ffc82a8e6.
- Speechmaticspricesource ↗
Speechmatics Batch Standard pre-recorded transcription price is unchanged
A correction, superseding the row it replaces.
Speechmatics lists Batch Standard pre-recorded transcription at $0.45 an hour ($0.0075 a minute) on pay as you go, which its pricing page calls the listed hourly rate with no discounts. The page's own data gave the same rate on 10 September, when the record first read it, so no price moved on 26 October. Until then the record held $0.24 an hour, the figure the page's table printed after an opt-in 33 percent model-training discount and a 20 percent volume discount that applies only above 500 hours a month; on 26 October it read the undiscounted rate for the first time.
- Speechmaticsprice · detectedsource ↗
Speechmatics Batch Melia 1 pre-recorded transcription rises to $0.00
Corrected. The row that supersedes this one stands in the record.
Pre-recorded transcription for Speechmatics Batch Melia 1 moves from $0.00 to $0.00 in USD per audio-minute, a 85 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 48d3d8603d1a and 753ffc82a8e6.
- Speechmaticspricesource ↗
Speechmatics Batch Melia 1 pre-recorded transcription price is unchanged
A correction, superseding the row it replaces.
Speechmatics lists Batch Melia 1 pre-recorded transcription at $0.24 an hour ($0.004 a minute) on pay as you go, which its pricing page calls the listed hourly rate with no discounts. The page's own data gave the same rate on 10 September, when the record first read it, so no price moved on 26 October. Until then the record held $0.13 an hour, the figure the page's table printed after an opt-in 33 percent model-training discount and a 20 percent volume discount that applies only above 500 hours a month; on 26 October it read the undiscounted rate for the first time.
- Nebiusprice · detectedsource ↗
Nebius B300 on-demand list price rises to $9.50
Corrected. The row that supersedes this one stands in the record.
On-demand list price for Nebius B300 moves from $7.85 to $9.50 in USD per GPU-hour, a 21 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 52edd8b0e197 and db6d79e914d6.
- Nebiuspricesource ↗
Nebius B300 on-demand list price rose 21 percent on 1 October
A correction, superseding the row it replaces.
Nebius raised the on-demand list price of an HGX B300 GPU from $7.85 to $9.50 an hour, a 21 percent rise, on 1 October 2026. Its pricing page announced the new rate in a column headed "GPU-hour (Effective October 1, 2026)", which the record archived on 22 September as 9a5f3542991c. The record dates the move 26 October because its reads of the page failed from 30 September until then, so it first read the new rate that day, already in force.
- Nebiusprice · detectedsource ↗
Nebius B200 on-demand list price rises to $8.50
Corrected. The row that supersedes this one stands in the record.
On-demand list price for Nebius B200 moves from $7.15 to $8.50 in USD per GPU-hour, a 19 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 52edd8b0e197 and db6d79e914d6.
- Nebiuspricesource ↗
Nebius B200 on-demand list price rose 19 percent on 1 October
A correction, superseding the row it replaces.
Nebius raised the on-demand list price of an HGX B200 GPU from $7.15 to $8.50 an hour, a 19 percent rise, on 1 October 2026. Its pricing page announced the new rate in a column headed "GPU-hour (Effective October 1, 2026)", which the record archived on 22 September as 9a5f3542991c. The record dates the move 26 October because its reads of the page failed from 30 September until then, so it first read the new rate that day, already in force.
- Nebiusprice · detectedsource ↗
Nebius H200 on-demand list price rises to $5.40
Corrected. The row that supersedes this one stands in the record.
On-demand list price for Nebius H200 moves from $4.50 to $5.40 in USD per GPU-hour, a 20 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 52edd8b0e197 and db6d79e914d6.
- Nebiuspricesource ↗
Nebius H200 on-demand list price rose 20 percent on 1 October
A correction, superseding the row it replaces.
Nebius raised the on-demand list price of an HGX H200 GPU from $4.50 to $5.40 an hour, a 20 percent rise, on 1 October 2026. Its pricing page announced the new rate in a column headed "GPU-hour (Effective October 1, 2026)", which the record archived on 22 September as 9a5f3542991c. The record dates the move 26 October because its reads of the page failed from 30 September until then, so it first read the new rate that day, already in force.
- Nebiusprice · detectedsource ↗
Nebius H100 SXM on-demand list price rises to $4.50
Corrected. The row that supersedes this one stands in the record.
On-demand list price for Nebius H100 SXM moves from $3.85 to $4.50 in USD per GPU-hour, a 17 percent rise. Read on 26 Oct 2026 at 26 Oct 04:02 UTC; the documents before and after are archived as 52edd8b0e197 and db6d79e914d6.
- Nebiuspricesource ↗
Nebius H100 SXM on-demand list price rose 17 percent on 1 October
A correction, superseding the row it replaces.
Nebius raised the on-demand list price of an HGX H100 GPU from $3.85 to $4.50 an hour, a 17 percent rise, on 1 October 2026. Its pricing page announced the new rate in a column headed "GPU-hour (Effective October 1, 2026)", which the record archived on 22 September as 9a5f3542991c. The record dates the move 26 October because its reads of the page failed from 30 September until then, so it first read the new rate that day, already in force.
- Browserbaseprice · detectedsource ↗
Browserbase Search API per 1,000 calls falls to $5.00
Corrected. The row that supersedes this one stands in the record.
Per 1,000 calls for Browserbase Search API moves from $7.00 to $5.00 in USD per 1,000 calls, a 29 percent cut. Read on 26 Oct 2026 at 26 Oct 04:01 UTC; the documents before and after are archived as 8278e33db3a6 and c5518bf3b4b9.
- Browserbasepricesource ↗
Browserbase Search API now prices three search backends apart
A correction, superseding the row it replaces.
Browserbase's Developer plan card prices search at "$5/1k searches by default with more options", where the card the record read on 14 October said "$7/1k calls". The plan comparison now prices three search backends apart on the Developer and Startup plans: Parallel Fast at $1, Parallel Advanced at $5 and Exa at $7 per 1,000 calls. The page does not say which backend is the default, so whether searches got cheaper or the default moved to a cheaper backend is not on it, and $7 is now Exa's rate. The record dates the change 26 October, its first good read of the page after 14 October, so the page changed at some point between those two reads.
- RunPodprice · detectedsource ↗
RunPod B200 on-demand list price rises to $7.99
On-demand list price for RunPod B200 moves from $6.79 to $7.99 in USD per GPU-hour, a 18 percent rise. Read on 15 Oct 2026 at 15 Oct 14:04 UTC; the documents before and after are archived as 49f205178f49 and d90fa84891af.
- RunPodprice · detectedsource ↗
RunPod H200 on-demand list price rises to $5.29
On-demand list price for RunPod H200 moves from $4.59 to $5.29 in USD per GPU-hour, a 15 percent rise. Read on 15 Oct 2026 at 15 Oct 14:04 UTC; the documents before and after are archived as 49f205178f49 and d90fa84891af.
- RunPodprice · detectedsource ↗
RunPod H100 SXM on-demand list price rises to $3.99
On-demand list price for RunPod H100 SXM moves from $3.49 to $3.99 in USD per GPU-hour, a 14 percent rise. Read on 15 Oct 2026 at 15 Oct 14:04 UTC; the documents before and after are archived as 49f205178f49 and d90fa84891af.
- Together AIprice · detectedsource ↗
Together AI Kimi K3 cached input price rises to $0.30
Cached input price for Together AI Kimi K3 moves from $0.27 to $0.30 in USD per 1M tokens, a 11 percent rise. Read on 13 Oct 2026 at 13 Oct 17:03 UTC; the documents before and after are archived as 035068744527 and 92f5eb2bf9ca.
- Together AIprice · detectedsource ↗
Together AI Kimi K3 output price rises to $15.00
Output price for Together AI Kimi K3 moves from $13.50 to $15.00 in USD per 1M tokens, a 11 percent rise. Read on 12 Oct 2026 at 12 Oct 07:08 UTC; the documents before and after are archived as f535a19de9b1 and 035068744527.
- Together AIprice · detectedsource ↗
Together AI Kimi K3 input price rises to $3.00
Input price for Together AI Kimi K3 moves from $2.70 to $3.00 in USD per 1M tokens, a 11 percent rise. Read on 12 Oct 2026 at 12 Oct 07:08 UTC; the documents before and after are archived as f535a19de9b1 and 035068744527.
- DeepInfraprice · detectedsource ↗
DeepInfra Mistral-Nemo-Instruct-2407 input price rises to $0.03
Input price for DeepInfra Mistral-Nemo-Instruct-2407 moves from $0.02 to $0.03 in USD per 1M tokens, a 53 percent rise. Read on 8 Oct 2026 at 8 Oct 20:08 UTC; the documents before and after are archived as 0075e943736d and cd498c0b23df.
- Anthropicprice · detectedsource ↗
Anthropic Claude Sonnet 5.5 cached input price falls to $0.10
Cached input price for Anthropic Claude Sonnet 5.5 moves from $0.20 to $0.10 in USD per 1M tokens, a 50 percent cut. Read on 7 Oct 2026 at 7 Oct 22:00 UTC; the documents before and after are archived as 9c0140eac6b4 and 6b5f41edbdbd.
- Together AIprice · detectedsource ↗
Together AI Kimi K3 output price falls to $13.50
Output price for Together AI Kimi K3 moves from $15.00 to $13.50 in USD per 1M tokens, a 10 percent cut. Read on 5 Oct 2026 at 5 Oct 23:00 UTC; the documents before and after are archived as caaa7b0c67f9 and c27ff13b44ce.
- Together AIprice · detectedsource ↗
Together AI Kimi K3 cached input price falls to $0.27
Cached input price for Together AI Kimi K3 moves from $0.30 to $0.27 in USD per 1M tokens, a 10 percent cut. Read on 5 Oct 2026 at 5 Oct 23:00 UTC; the documents before and after are archived as caaa7b0c67f9 and c27ff13b44ce.
- Together AIprice · detectedsource ↗
Together AI Kimi K3 input price falls to $2.70
Input price for Together AI Kimi K3 moves from $3.00 to $2.70 in USD per 1M tokens, a 10 percent cut. Read on 5 Oct 2026 at 5 Oct 23:00 UTC; the documents before and after are archived as caaa7b0c67f9 and c27ff13b44ce.
- Together AIprice · detectedsource ↗
Together AI Qwen3.8-Flash output price rises to $0.47
Output price for Together AI Qwen3.8-Flash moves from $0.28 to $0.47 in USD per 1M tokens, a 68 percent rise. Read on 5 Oct 2026 at 5 Oct 23:00 UTC; the documents before and after are archived as caaa7b0c67f9 and c27ff13b44ce.
- Together AIprice · detectedsource ↗
Together AI Qwen3.8-Flash input price rises to $0.15
Input price for Together AI Qwen3.8-Flash moves from $0.09 to $0.15 in USD per 1M tokens, a 67 percent rise. Read on 5 Oct 2026 at 5 Oct 23:00 UTC; the documents before and after are archived as caaa7b0c67f9 and c27ff13b44ce.
- DeepInfraprice · detectedsource ↗
DeepInfra gemma-4-31B-it input price rises to $0.20
Input price for DeepInfra gemma-4-31B-it moves from $0.15 to $0.20 in USD per 1M tokens, a 33 percent rise. Read on 4 Oct 2026 at 4 Oct 21:06 UTC; the documents before and after are archived as 10659a3fe934 and ecb461ecf066.
- DeepInfraprice · detectedsource ↗
DeepInfra Qwen3.6-35B-A3B output price falls to $0.10
Output price for DeepInfra Qwen3.6-35B-A3B moves from $0.95 to $0.10 in USD per 1M tokens, a 89 percent cut. Read on 2 Oct 2026 at 2 Oct 10:06 UTC; the documents before and after are archived as 621fade6384b and 814a5b1e1469.
- DeepInfraprice · detectedsource ↗
DeepInfra gemma-4-31B-it output price rises to $0.40
Output price for DeepInfra gemma-4-31B-it moves from $0.38 to $0.40 in USD per 1M tokens, a 5 percent rise. Read on 1 Oct 2026 at 1 Oct 20:09 UTC; the documents before and after are archived as f2cc8f57c2ff and ea7043e7e79b.
- DeepInfraprice · detectedsource ↗
DeepInfra gemma-4-31B-it input price rises to $0.15
Input price for DeepInfra gemma-4-31B-it moves from $0.13 to $0.15 in USD per 1M tokens, a 15 percent rise. Read on 1 Oct 2026 at 1 Oct 20:09 UTC; the documents before and after are archived as f2cc8f57c2ff and ea7043e7e79b.
- Speechmaticsprice · detectedsource ↗
Speechmatics Batch Melia 1 pre-recorded transcription rises to $0.00
Corrected. The row that supersedes this one stands in the record.
Pre-recorded transcription for Speechmatics Batch Melia 1 moves from $0.00 to $0.00 in USD per audio-minute, a 1 percent rise. Read on 30 Sep 2026 at 30 Sep 13:02 UTC; the documents before and after are archived as 58999a3aefb7 and fc1c5cb62099.
- Speechmaticspricesource ↗
Speechmatics Batch Melia 1 pre-recorded transcription price is unchanged
A correction, superseding the row it replaces.
Speechmatics lists Batch Melia 1 pre-recorded transcription at $0.24 an hour ($0.004 a minute) on pay as you go, the rate its pricing page's own data gave on 10 September and again on 26 October, so no price moved on 30 September. The record then held the figure the page's table printed after an opt-in 33 percent model-training discount and a 20 percent volume discount that applies only above 500 hours a month, and that figure went from $0.129 to $0.13 an hour: $0.24 less both discounts is $0.12864, and the two are that one figure printed to three places and then to two.
- DeepInfraprice · detectedsource ↗
DeepInfra NVIDIA-Nemotron-3.5-Lightning output price falls to $0.16
Output price for DeepInfra NVIDIA-Nemotron-3.5-Lightning moves from $0.20 to $0.16 in USD per 1M tokens, a 20 percent cut. Read on 29 Sep 2026 at 29 Sep 00:04 UTC; the documents before and after are archived as d361986b2059 and baa29b806ed5.
- DeepInfraprice · detectedsource ↗
DeepInfra NVIDIA-Nemotron-3.5-Lightning cached input price falls to $0.03
Cached input price for DeepInfra NVIDIA-Nemotron-3.5-Lightning moves from $0.04 to $0.03 in USD per 1M tokens, a 25 percent cut. Read on 29 Sep 2026 at 29 Sep 00:04 UTC; the documents before and after are archived as d361986b2059 and baa29b806ed5.
- DeepInfraprice · detectedsource ↗
DeepInfra NVIDIA-Nemotron-3.5-Lightning input price falls to $0.06
Input price for DeepInfra NVIDIA-Nemotron-3.5-Lightning moves from $0.08 to $0.06 in USD per 1M tokens, a 25 percent cut. Read on 29 Sep 2026 at 29 Sep 00:04 UTC; the documents before and after are archived as d361986b2059 and baa29b806ed5.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max output price falls to $4.50
Output price for Together AI Qwen3.7-Max moves from $7.50 to $4.50 in USD per 1M tokens, a 40 percent cut. Read on 22 Sep 2026 at 22 Sep 22:02 UTC; the documents before and after are archived as 5d3cbeab14ac and 4859579db281.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max cached input price rises to $0.30
Cached input price for Together AI Qwen3.7-Max moves from $0.25 to $0.30 in USD per 1M tokens, a 20 percent rise. Read on 22 Sep 2026 at 22 Sep 22:02 UTC; the documents before and after are archived as 5d3cbeab14ac and 4859579db281.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max input price falls to $1.50
Input price for Together AI Qwen3.7-Max moves from $2.50 to $1.50 in USD per 1M tokens, a 40 percent cut. Read on 22 Sep 2026 at 22 Sep 22:02 UTC; the documents before and after are archived as 5d3cbeab14ac and 4859579db281.
- Together AIprice · detectedsource ↗
Together AI Qwen3.8-Flash output price falls to $0.28
Output price for Together AI Qwen3.8-Flash moves from $0.47 to $0.28 in USD per 1M tokens, a 40 percent cut. Read on 22 Sep 2026 at 22 Sep 22:02 UTC; the documents before and after are archived as 5d3cbeab14ac and 4859579db281.
- Together AIprice · detectedsource ↗
Together AI Qwen3.8-Flash input price falls to $0.09
Input price for Together AI Qwen3.8-Flash moves from $0.15 to $0.09 in USD per 1M tokens, a 40 percent cut. Read on 22 Sep 2026 at 22 Sep 22:02 UTC; the documents before and after are archived as 5d3cbeab14ac and 4859579db281.
- Anthropiclaunchsource ↗
Claude Opus 5.5 released
Succeeds Opus 5 at $4 / $20, a list price 20 percent below Opus 5's $5 / $25, with cache reads at 0.05x base input rather than 0.1x.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max output price rises to $7.50
Output price for Together AI Qwen3.7-Max moves from $6.00 to $7.50 in USD per 1M tokens, a 25 percent rise. Read on 16 Sep 2026 at 16 Sep 05:04 UTC; the documents before and after are archived as a751e17fa0e6 and 015a2bd08acd.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max input price rises to $2.50
Input price for Together AI Qwen3.7-Max moves from $2.00 to $2.50 in USD per 1M tokens, a 25 percent rise. Read on 16 Sep 2026 at 16 Sep 05:04 UTC; the documents before and after are archived as a751e17fa0e6 and 015a2bd08acd.
- Together AIlaunchsource ↗
DeepSeek V4.1 Flash listed on serving pages
Corrected. The row that supersedes this one stands in the record.
DeepSeek V4.1 Flash appears on serverless serving pages, priced identically across providers at $0.15 input and $0.30 output per million tokens.
- Together AIlaunchsource ↗
DeepSeek V4.1 Flash is listed on serving pages
A correction, superseding the row it replaces.
DeepSeek V4.1 Flash appears on Together AI's serverless pricing page at $0.30 input, $0.006 cached input and $1.20 output per million tokens.
- Azureprice · detectedsource ↗
Azure GPT-5.6 Sol output price falls to $20.00
Output price for Azure GPT-5.6 Sol moves from $30.00 to $20.00 in USD per 1M tokens, a 33 percent cut. Read on 11 Sep 2026 at 11 Sep 00:07 UTC; the documents before and after are archived as fd08758b1b41 and ed6bae55a487.
- Azureprice · detectedsource ↗
Azure GPT-5.6 Sol cached input price falls to $0.40
Cached input price for Azure GPT-5.6 Sol moves from $0.50 to $0.40 in USD per 1M tokens, a 20 percent cut. Read on 11 Sep 2026 at 11 Sep 00:07 UTC; the documents before and after are archived as fd08758b1b41 and ed6bae55a487.
- Azureprice · detectedsource ↗
Azure GPT-5.6 Sol input price falls to $4.00
Input price for Azure GPT-5.6 Sol moves from $5.00 to $4.00 in USD per 1M tokens, a 20 percent cut. Read on 11 Sep 2026 at 11 Sep 00:07 UTC; the documents before and after are archived as fd08758b1b41 and ed6bae55a487.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max output price rises to $6.00
Output price for Together AI Qwen3.7-Max moves from $3.75 to $6.00 in USD per 1M tokens, a 60 percent rise. Read on 10 Sep 2026 at 10 Sep 17:09 UTC; the documents before and after are archived as 8911c9c9a47d and 0810329b8902.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max cached input price rises to $0.25
Cached input price for Together AI Qwen3.7-Max moves from $0.13 to $0.25 in USD per 1M tokens, a 92 percent rise. Read on 10 Sep 2026 at 10 Sep 17:09 UTC; the documents before and after are archived as 8911c9c9a47d and 0810329b8902.
- Together AIprice · detectedsource ↗
Together AI Qwen3.7-Max input price rises to $2.00
Input price for Together AI Qwen3.7-Max moves from $1.25 to $2.00 in USD per 1M tokens, a 60 percent rise. Read on 10 Sep 2026 at 10 Sep 17:09 UTC; the documents before and after are archived as 8911c9c9a47d and 0810329b8902.
- Novita AIavailabilitysource ↗
Novita withdraws DeepSeek V3 0324, V3 Turbo, and Kat Coder Pro
Novita AI removes three legacy serving configurations from its public endpoints following the availability of V3.1 and V3.2 variants.
- Novita AIprice · detectedsource ↗
Novita AI Macaron V1 Tall output price falls to $1.43
Output price for Novita AI Macaron V1 Tall moves from $2.60 to $1.43 in USD per 1M tokens, a 45 percent cut. Read on 9 Sep 2026 at 9 Sep 18:03 UTC; the documents before and after are archived as 2e479a186ccf and 2af544f9f12f.
- Novita AIprice · detectedsource ↗
Novita AI Macaron V1 Tall cached input price falls to $0.04
Cached input price for Novita AI Macaron V1 Tall moves from $0.08 to $0.04 in USD per 1M tokens, a 45 percent cut. Read on 9 Sep 2026 at 9 Sep 18:03 UTC; the documents before and after are archived as 2e479a186ccf and 2af544f9f12f.
- Novita AIprice · detectedsource ↗
Novita AI Macaron V1 Tall input price falls to $0.25
Input price for Novita AI Macaron V1 Tall moves from $0.45 to $0.25 in USD per 1M tokens, a 45 percent cut. Read on 9 Sep 2026 at 9 Sep 18:03 UTC; the documents before and after are archived as 2e479a186ccf and 2af544f9f12f.
- Novita AIprice · detectedsource ↗
Novita AI Macaron V1 Venti output price falls to $2.48
Output price for Novita AI Macaron V1 Venti moves from $4.50 to $2.48 in USD per 1M tokens, a 45 percent cut. Read on 9 Sep 2026 at 9 Sep 18:03 UTC; the documents before and after are archived as 2e479a186ccf and 2af544f9f12f.
- Novita AIprice · detectedsource ↗
Novita AI Macaron V1 Venti cached input price falls to $0.17
Cached input price for Novita AI Macaron V1 Venti moves from $0.30 to $0.17 in USD per 1M tokens, a 45 percent cut. Read on 9 Sep 2026 at 9 Sep 18:03 UTC; the documents before and after are archived as 2e479a186ccf and 2af544f9f12f.
- Novita AIprice · detectedsource ↗
Novita AI Macaron V1 Venti input price falls to $0.82
Input price for Novita AI Macaron V1 Venti moves from $1.50 to $0.82 in USD per 1M tokens, a 45 percent cut. Read on 9 Sep 2026 at 9 Sep 18:03 UTC; the documents before and after are archived as 2e479a186ccf and 2af544f9f12f.
- Novita AIprice · detectedsource ↗
Novita AI DeepSeek V4 Pro 0813 cached input price falls to $0.04
Cached input price for Novita AI DeepSeek V4 Pro 0813 moves from $0.13 to $0.04 in USD per 1M tokens, a 67 percent cut. Read on 9 Sep 2026 at 9 Sep 18:03 UTC; the documents before and after are archived as 2e479a186ccf and 2af544f9f12f.
- Together AIavailabilitysource ↗
Together delists older DeepSeek and Gemma variants
Together AI unlists older serving tiers and preview checkpoints across its serverless serving catalog.
- Googleavailabilitysource ↗
Google withdraws Gemini 2.0 Flash from API pricing
Gemini 2.0 Flash is retired from standard production billing in favour of Gemini 2.5 Flash and Flash-Lite variants.
- Googleprice · detectedsource ↗
Google Gemini Robotics ER 2 Streaming Preview output price falls to $5.00
Output price for Google Gemini Robotics ER 2 Streaming Preview moves from $10.00 to $5.00 in USD per 1M tokens, a 50 percent cut. Read on 8 Sep 2026 at 8 Sep 16:04 UTC; the documents before and after are archived as 2c6e21cae722 and b7ced7746bfc.
- Googleprice · detectedsource ↗
Google Gemini Robotics ER 2 Streaming Preview input price falls to $1.00
Input price for Google Gemini Robotics ER 2 Streaming Preview moves from $2.00 to $1.00 in USD per 1M tokens, a 50 percent cut. Read on 8 Sep 2026 at 8 Sep 16:04 UTC; the documents before and after are archived as 2c6e21cae722 and b7ced7746bfc.
- Googleprice · detectedsource ↗
Google Gemini Robotics ER 2 Preview batch output price falls to $2.50
Batch output price for Google Gemini Robotics ER 2 Preview moves from $5.00 to $2.50 in USD per 1M tokens, a 50 percent cut. Read on 8 Sep 2026 at 8 Sep 16:04 UTC; the documents before and after are archived as 2c6e21cae722 and b7ced7746bfc.
- Googleprice · detectedsource ↗
Google Gemini Robotics ER 2 Preview batch input price falls to $0.50
Batch input price for Google Gemini Robotics ER 2 Preview moves from $1.00 to $0.50 in USD per 1M tokens, a 50 percent cut. Read on 8 Sep 2026 at 8 Sep 16:04 UTC; the documents before and after are archived as 2c6e21cae722 and b7ced7746bfc.
- Googleprice · detectedsource ↗
Google Gemini Robotics ER 2 Preview output price falls to $5.00
Output price for Google Gemini Robotics ER 2 Preview moves from $10.00 to $5.00 in USD per 1M tokens, a 50 percent cut. Read on 8 Sep 2026 at 8 Sep 16:04 UTC; the documents before and after are archived as 2c6e21cae722 and b7ced7746bfc.
- Googleprice · detectedsource ↗
Google Gemini Robotics ER 2 Preview cached input price falls to $0.10
Cached input price for Google Gemini Robotics ER 2 Preview moves from $0.20 to $0.10 in USD per 1M tokens, a 50 percent cut. Read on 8 Sep 2026 at 8 Sep 16:04 UTC; the documents before and after are archived as 2c6e21cae722 and b7ced7746bfc.
- Googleprice · detectedsource ↗
Google Gemini Robotics ER 2 Preview input price falls to $1.00
Input price for Google Gemini Robotics ER 2 Preview moves from $2.00 to $1.00 in USD per 1M tokens, a 50 percent cut. Read on 8 Sep 2026 at 8 Sep 16:04 UTC; the documents before and after are archived as 2c6e21cae722 and b7ced7746bfc.
- DeepInfraprice · detectedsource ↗
DeepInfra DeepSeek V4 Flash 0731 cached input price falls to $0.01
Cached input price for DeepInfra DeepSeek V4 Flash 0731 moves from $0.02 to $0.01 in USD per 1M tokens, a 6 percent cut. Read on 6 Sep 2026 at 6 Sep 08:07 UTC; the documents before and after are archived as 7aa861afcec3 and 3d78a957bf2f.
- DeepInfraprice · detectedsource ↗
DeepInfra DeepSeek V4 Flash 0731 input price falls to $0.06
Input price for DeepInfra DeepSeek V4 Flash 0731 moves from $0.08 to $0.06 in USD per 1M tokens, a 25 percent cut. Read on 6 Sep 2026 at 6 Sep 08:07 UTC; the documents before and after are archived as 7aa861afcec3 and 3d78a957bf2f.
- RunPodprice · detectedsource ↗
RunPod H100 SXM on-demand list price rises to $3.49
On-demand list price for RunPod H100 SXM moves from $3.29 to $3.49 in USD per GPU-hour, a 6 percent rise. Read on 6 Sep 2026 at 6 Sep 01:04 UTC; the documents before and after are archived as c03a0ac34b88 and f4ebcd1247e1.
- DeepInfraprice · detectedsource ↗
DeepInfra Claude Sonnet 5 output price rises to $15.00
Output price for DeepInfra Claude Sonnet 5 moves from $10.00 to $15.00 in USD per 1M tokens, a 50 percent rise. Read on 1 Sep 2026 at 1 Sep 08:08 UTC; the documents before and after are archived as dfa732ec2a2d and 595d2e3c602a.
- DeepInfraprice · detectedsource ↗
DeepInfra Claude Sonnet 5 input price rises to $3.00
Input price for DeepInfra Claude Sonnet 5 moves from $2.00 to $3.00 in USD per 1M tokens, a 50 percent rise. Read on 1 Sep 2026 at 1 Sep 08:08 UTC; the documents before and after are archived as dfa732ec2a2d and 595d2e3c602a.
- Together AIprice · detectedsource ↗
Together AI Qwen3.8-2.4T-A95B output price falls to $6.00
Output price for Together AI Qwen3.8-2.4T-A95B moves from $6.25 to $6.00 in USD per 1M tokens, a 4 percent cut. Read on 28 Aug 2026 at 28 Aug 18:07 UTC; the documents before and after are archived as 8404e6f47b8a and abff4ef9eb2f.
- Together AIprice · detectedsource ↗
Together AI Qwen3.8-2.4T-A95B cached input price falls to $0.25
Cached input price for Together AI Qwen3.8-2.4T-A95B moves from $0.50 to $0.25 in USD per 1M tokens, a 50 percent cut. Read on 28 Aug 2026 at 28 Aug 18:07 UTC; the documents before and after are archived as 8404e6f47b8a and abff4ef9eb2f.
- Together AIprice · detectedsource ↗
Together AI Qwen3.8-2.4T-A95B input price falls to $2.00
Input price for Together AI Qwen3.8-2.4T-A95B moves from $2.50 to $2.00 in USD per 1M tokens, a 20 percent cut. Read on 28 Aug 2026 at 28 Aug 18:07 UTC; the documents before and after are archived as 8404e6f47b8a and abff4ef9eb2f.
- Novita AIavailabilitysource ↗
Novita drops the Ling and Ring 2.6 models
Ling 2.6 1T, Ling 2.6 Flash and Ring 2.6 1T are no longer listed. Ling 3.0 Flash and Ling 3.0 Flash Fast are, both at $0.06 / $0.18 per million tokens, and the three withdrawn rows are off the boards. Vitna's own reads of the page bracket the removal to this date; how long the 3.0 models had been listed before it is not something this record can say, because it first read this page on 22 August.
- OpenAIpricesource ↗
GPT-5.6 Sol cut 20 percent on input and 33 percent on output
The frontier tier follows Luna and Terra down, six weeks after the generation launched.
- Googlelaunchsource ↗
Gemini 3.7 Flash reaches general availability
The workhorse tier steps up, listing at $0.75 / $3.75 per million tokens through December 31, then $1.50 / $7.50.
- Anthropicpricesource ↗
Claude Sonnet 5 introductory price becomes the standard price
The $2 / $10 per million tokens announced at launch was to rise to $3 / $15 on 1 September. That increase will not occur.
- OpenAIavailabilitysource ↗
Daybreak Blue and Daybreak Red open to approved defenders
Two access tiers released to vetted customers rather than to the API generally.
- Anthropicpolicysource ↗
Claude Opus 4.1 retired
Corrected. The row that supersedes this one stands in the record.
Deprecated June 5 with 60 days' notice; requests to the model now fail. The recommended replacement is Opus 4.8.
- Anthropicavailabilitysource ↗
Claude Opus 4.1 is retired
A correction, superseding the row it replaces.
Corrects the kind on the row this supersedes, which was entered as policy: a retirement is an availability event, and it is how every other retirement in this record is filed. Deprecated 5 June with the sixty days of notice the policy promises; the recommended replacement is Opus 4.8. The dates on the cited page bind Anthropic-operated platforms, and Amazon Bedrock and Google Cloud set their own retirement schedules.
- OpenAIpricesource ↗
GPT-5.6 Terra list price cut 20 percent
The balanced tier is cut on the same day as Luna, and Fast mode replaces Priority processing.
- OpenAIpricesource ↗
GPT-5.6 Luna list price cut about 80 percent
Three weeks after launch, Luna drops from $1.00 / $6.00 to $0.20 / $1.20 per million tokens.
- Anthropiclaunchsource ↗
Claude Opus 5 released
Succeeds Opus 4.5 at the same $5 / $25 list price.
- Googlelaunchsource ↗
Gemini 3.6 Flash and 3.5 Flash-Lite reach general availability
A new fast tier and a low-latency, cost-focused variant land on the same day.
- OpenAIdeprecationsource ↗
Nine legacy audio and realtime models are deprecated
Scheduled for 20 January 2027, with GPT-Realtime-2.1 and GPT-Audio-1.5 named as the replacement.
- OpenAIlaunchsource ↗
GPT-5.6 Sol, Terra and Luna launch
Luna, the fast tier, lists at $1.00 / $6.00 per million tokens.
- Anthropiclaunchsource ↗
Claude Sonnet 5 released
Opus-class coding and tool use at a $2 / $10 list price.
- Anthropicavailabilitysource ↗
Claude Sonnet 4 and Claude Opus 4 are retired
Requests to both now fail on the Claude API. Each remains served on at least one partner cloud.
- Anthropicpolicysource ↗
US export-control directive suspends Fable 5 and Mythos 5
All customer access disabled three days after launch; restored at the end of June.
- OpenAIdeprecationsource ↗
The GPT-5 and o3 snapshots are deprecated
Six models scheduled for 11 December 2026, sixteen months after the GPT-5 family launched.
- Anthropiclaunchsource ↗
Claude Fable 5 opens the Mythos tier
The first publicly available model above the Opus class, at $10 / $50.
- Anthropicdeprecationsource ↗
Claude Opus 4.1 is deprecated
Retirement set for 5 August 2026, the sixty days of notice the policy promises.
- Googleavailabilitysource ↗
Four Gemini 2.0 models shut down
The shutdown was announced on 18 February, one hundred and three days ahead.
- Anthropiclaunchsource ↗
Claude Opus 4.8 released
A 1M token context window by default across the Claude API, Bedrock, Google Cloud and Microsoft Foundry.
- Googlelaunchsource ↗
Gemini 3.5 Flash reaches general availability
Managed Agents opens in public preview on the same day.
- Googlelaunchsource ↗
Gemini 3.1 Flash-Lite reaches general availability
The preview it replaces is scheduled for shutdown eighteen days later.
- OpenAIlaunchsource ↗
GPT-5.5 and GPT-5.5 Pro reach a million tokens
Both support a 1M token context window.
- OpenAIdeprecationsource ↗
Thirteen legacy GPT snapshots are deprecated at once
Scheduled for 23 October 2026, with the three GPT-5.6 tiers named as the replacement.
- Anthropicavailabilitysource ↗
Claude Haiku 3 is retired
Requests now fail, twenty-five months after the model launched with the Claude 3 family.
- Anthropiclaunchsource ↗
Claude Opus 4.7 released at the price of Opus 4.6
The same $5 / $25 per million tokens, on a new tokenizer that produces roughly 30 percent more tokens for the same text.
- Anthropicdeprecationsource ↗
Claude Sonnet 4 and Claude Opus 4 are deprecated
Retirement set for 15 June 2026 for both, sixty-two days of notice.
- Anthropicavailabilitysource ↗
Claude Mythos Preview opens by invitation only
A gated research preview for defensive cybersecurity work under Project Glasswing. There is no self-serve sign-up.
- Googlepolicysource ↗
Flex and Priority inference tiers open on the Gemini API
Two service levels at two prices for the same models, the latency-for-money trade the batch discount makes in reverse.
- OpenAIlaunchsource ↗
GPT-5.4 mini and nano complete the ladder
The small and smallest tiers land twelve days after the frontier pair.
- Anthropicpolicysource ↗
The 1M token context window leaves beta at standard pricing
Requests over 200k tokens work with no beta header on Opus 4.6 and Sonnet 4.6, and are billed at the same per-token rate as a short one.
- OpenAIlaunchsource ↗
GPT-5.4 and GPT-5.4 Pro released
Tool search, computer use and compaction arrive with them.
- Googlelaunchsource ↗
Gemini 3.1 Pro enters preview
The first 3.1 release, three months after the 3 generation opened.
- Anthropicavailabilitysource ↗
Claude Sonnet 3.7 and Claude Haiku 3.5 are retired
Both stop answering on the Claude API on the same day. Haiku 3.5 remains served on Amazon Bedrock and Google Cloud.
- Anthropicdeprecationsource ↗
Claude Haiku 3 is deprecated
Retirement set for 20 April 2026, announced on the day two other models are retired.
- Anthropiclaunchsource ↗
Claude Sonnet 4.6 released
The balanced tier gains the 1M token context window and better agentic search at the same $3 / $15.
- Anthropicpolicysource ↗
US-only inference is priced at a premium
Data residency controls arrive with the inference_geo parameter; pinning inference to the US costs 1.1x standard rates on models released after 1 February 2026.
- Anthropiclaunchsource ↗
Claude Opus 4.6 released
Recommends adaptive thinking, and opens the 1M token context window in beta on the Opus line.
- Anthropicavailabilitysource ↗
Claude Opus 3 is retired
Requests now fail, more than six months after the notice and twenty-two months after the model launched.
- 2025Anthropicdeprecationsource ↗
Claude Haiku 3.5 is deprecated
Retirement set for 19 February 2026, the same day Sonnet 3.7 goes.
- Googlelaunchsource ↗
Gemini 3 Flash enters preview
The fast tier of the 3 generation, with multimodal function responses.
- OpenAIlaunchsource ↗
GPT-5.2 released with an extra reasoning step
Adds an xhigh reasoning effort and client-side compaction.
- Anthropiclaunchsource ↗
Claude Opus 4.5 released
The frontier tier moves to $5 / $25 per million tokens.
- Googlelaunchsource ↗
Gemini 3 Pro enters preview
The 3 generation opens with media resolution and thinking-level controls.
- OpenAIlaunchsource ↗
GPT-5.1 released with a codex pair beside it
Improved steerability and agentic work, with gpt-5.1-codex and gpt-5.1-codex-mini.
- Anthropicavailabilitysource ↗
The Claude Sonnet 3.5 snapshots are retired
Requests to both now fail, seventy-six days after the notice.
- Anthropicdeprecationsource ↗
Claude Sonnet 3.7 is deprecated on the day 3.5 goes
Retirement set for 19 February 2026, announced in the same release notes that retire its predecessor.
- Anthropiclaunchsource ↗
Claude Haiku 4.5 released
Near-frontier small-model quality at $1 / $5 per million tokens.
- OpenAIlaunchsource ↗
GPT-5 Pro tops the GPT-5 ladder
Released with a mini realtime tier, gpt-audio-mini, gpt-image-1-mini and the Sora 2 video API.
- Anthropicpolicysource ↗
Global endpoint pricing introduced on Bedrock and Vertex AI
Regional routing starts carrying a premium over global on the partner clouds. First-party Claude API pricing is unaffected.
- Anthropiclaunchsource ↗
Claude Sonnet 4.5 released
The balanced tier is refreshed for agents and coding at $3 / $15 per million tokens.
- OpenAIdeprecationsource ↗
GPT-3.5 Turbo Instruct and the base models are deprecated
Scheduled for 28 September 2026, with GPT-5.6 Terra named as the replacement for davinci-002, babbage-002 and the last GPT-3.5 snapshots.
- OpenAIdeprecationsource ↗
The Assistants API is deprecated
Scheduled for 26 August 2026, with the Responses and Conversations APIs named as the replacement. A year of notice for an interface rather than a model.
- Anthropicdeprecationsource ↗
Both Claude Sonnet 3.5 snapshots are deprecated
Retirement set for 28 October 2025 for the June and October 2024 snapshots.
- OpenAIlaunchsource ↗
The GPT-5 family launches with three tiers
gpt-5, gpt-5-mini and gpt-5-nano ship together, with a minimal reasoning effort setting.
- Anthropiclaunchsource ↗
Claude Opus 4.1 released
An incremental update to Opus 4 at the same frontier list price.
- Googlelaunchsource ↗
Gemini 2.5 Flash-Lite reaches general availability
The low-latency tier of the 2.5 generation leaves preview.
- Anthropicavailabilitysource ↗
Claude 2.0, Claude 2.1 and Claude Sonnet 3 are retired
Requests to all three now fail, on the date named in January.
- OpenAIavailabilitysource ↗
GPT-4.5 preview is shut down
Four and a half months of life from research preview to retirement, the shortest the record carries.
- Anthropicdeprecationsource ↗
Claude Opus 3 is deprecated
Retirement set for 5 January 2026, more than six months of notice.
- Googlelaunchsource ↗
Gemini 2.5 Pro and 2.5 Flash become stable
Both leave preview, and a 2.5 Flash-Lite preview opens under them.
- OpenAIpricesource ↗
o3 list price cut across every request type
The reduction lands on the same day o3-pro is released above it.
- Anthropiclaunchsource ↗
Claude Opus 4 and Claude Sonnet 4 launch together
The generation that introduces extended thinking across both tiers.
- OpenAIdeprecationsource ↗
o1-preview and o1-mini are deprecated
The models that opened the reasoning tier are scheduled for 28 July and 27 October 2025, nineteen months after o1-preview shipped.
- OpenAIlaunchsource ↗
o3 and o4-mini join the reasoning line
Two reasoning models released with the first Codex CLI.
- OpenAIdeprecationsource ↗
GPT-4.5 preview is deprecated on the day GPT-4.1 launches
The research preview is scheduled for removal in the same changelog entry that announces its replacement.
- OpenAIlaunchsource ↗
GPT-4.1 launches as a three-model ladder
gpt-4.1, gpt-4.1-mini and gpt-4.1-nano land on the same day.
- Googlelaunchsource ↗
Gemini 2.5 Pro arrives with thinking enabled
gemini-2.5-pro-exp-03-25 is the first Gemini released with thinking mode on by default.
- OpenAIlaunchsource ↗
GPT-4.5 enters research preview
A preview tier that is deprecated seven weeks later, the shortest life the record carries.
- Googlelaunchsource ↗
Gemini 2.0 Flash-Lite reaches general availability
A cost-focused variant twenty days after the tier above it.
- Anthropiclaunchsource ↗
Claude Sonnet 3.7 ships one model with two response modes
Near-instant answers or step-by-step extended thinking, selected per request rather than by choosing a model.
- Googlelaunchsource ↗
Gemini 2.0 Flash reaches general availability
gemini-2.0-flash-001 becomes the stable id, and a 2.0 Pro experimental lands beside it.
- OpenAIlaunchsource ↗
o3-mini opens a cheaper reasoning tier
The reasoning line gains a small model five months after o1-mini.
- Anthropicdeprecationsource ↗
Claude 2, Claude 2.1 and Claude Sonnet 3 are deprecated
Retirement set for 21 July 2025, six months of notice for three models at once.
- 2024OpenAIlaunchsource ↗
o1 reaches general availability with realtime and audio models
The reasoning tier leaves preview alongside new realtime and audio snapshots and WebRTC support.
- Googlelaunchsource ↗
Gemini 2.0 Flash enters public preview
The experimental release that opens the 2.0 generation.
- Anthropicavailabilitysource ↗
Claude 1 and the Instant models are retired
Requests to claude-1.0 through claude-instant-1.2 now fail, sixty-three days after the notice.
- Anthropiclaunchsource ↗
Claude Haiku 3.5 arrives as a text-only model
The fast tier is refreshed on the Claude API without vision at launch.
- Anthropiclaunchsource ↗
Claude Sonnet 3.5 is upgraded and gains computer use
A second Sonnet 3.5 snapshot ships with Anthropic-defined computer use tools.
- Anthropicpolicysource ↗
The Message Batches API opens at half the standard rate
Large asynchronous batches are processed for 50 percent less, matching the discount OpenAI set in April.
- OpenAIpolicysource ↗
Prompt caching, distillation and the Realtime API arrive together
Caching lands on the same day as the Realtime API, model distillation, image fine-tuning and Evals.
- OpenAIavailabilitysource ↗
The 2023 GPT-3.5 Turbo snapshots are shut down
gpt-3.5-turbo-0301, 0613 and the 16k variant stop answering on the same day, on the schedule set in June and November 2023.
- OpenAIlaunchsource ↗
o1-preview and o1-mini open the reasoning tier
A model class billed for the tokens it thinks with, priced apart from the GPT line.
- Anthropicdeprecationsource ↗
Claude 1 and the Instant models are deprecated
Retirement set for 6 November 2024. The first deprecation notice the record carries.
- Anthropicpolicysource ↗
Prompt caching enters beta on the Claude API
A cached prompt prefix is re-read at a fraction of the input rate, which changes the unit price of a repeated system prompt rather than the model.
- OpenAIlaunchsource ↗
GPT-4o mini opens the cost-efficient tier
A small omni model priced for volume; fine-tuning follows five days later.
- Anthropiclaunchsource ↗
Claude Sonnet 3.5 lands across three channels at once
Available on the Claude API, Amazon Bedrock and Vertex AI on the day it is announced.
- OpenAIavailabilitysource ↗
GPT-4 0314 is shut down
The first pinned GPT-4 snapshot stops answering, one year to the day after its notice.
- OpenAIdeprecationsource ↗
GPT-4 32k and the vision preview are deprecated
gpt-4-vision-preview is scheduled for 6 December 2024 and gpt-4-32k for a year later, after vision folded into the main model.
- Googlelaunchsource ↗
Gemini 1.5 Pro and 1.5 Flash reach general availability
The long-context pair leaves preview nine days after pay-as-you-go billing opened.
- OpenAIlaunchsource ↗
GPT-4o released in the API
The omni tier lands as a single model for text, vision and audio.
- OpenAIpolicysource ↗
The Batch API arrives and halves the rate for patient work
Asynchronous request processing at a discount to the synchronous rate, the first time latency is priced separately from capability.
- OpenAIlaunchsource ↗
GPT-4 Turbo with Vision reaches general availability
The preview tier becomes a generally available model with image input in the same request.
- Anthropiclaunchsource ↗
The Claude 3 family sets three price tiers at once
Opus, Sonnet and Haiku launch together at $15 / $75, $3 / $15 and $0.25 / $1.25 per million tokens, the first time one announcement names a whole ladder.
- OpenAIlaunchsource ↗
A dated GPT-3.5 Turbo snapshot joins the API
gpt-3.5-turbo-0125 is pinned as its own model id, and is still listed on the pricing index today.
- OpenAIavailabilitysource ↗
The pre-chat generation is shut down
InstructGPT, the base models, the edits endpoint and the first-generation embeddings stop answering, six months after the notice.
- 2023Googlelaunchsource ↗
The Gemini API opens to developers
Gemini Pro, Gemini Pro Vision, embedding-001 and aqa ship together, and the v1 stable channel opens with them.
- OpenAIlaunchsource ↗
GPT-4 Turbo opens the long-context tier
The first DevDay release: GPT-4 Turbo in preview alongside an updated GPT-3.5 Turbo, vision input, the Assistants API, DALL-E 3 and text to speech.
- OpenAIdeprecationsource ↗
GPT-3.5 Turbo 0613 is deprecated on the day GPT-4 Turbo launches
gpt-3.5-turbo-0613 and its 16k sibling are scheduled for 13 September 2024, announced in the same DevDay that opens the tier above them.
- Anthropiclaunchsource ↗
Claude 2 released at the price of Claude 1.3
A generation that improves on coding, maths and reasoning and does not move the rate.
- OpenAIdeprecationsource ↗
InstructGPT, the base models and the first embeddings are deprecated
text-davinci-003, the bare ada and davinci models, the edits endpoint and every first-generation embedding are scheduled for 4 January 2024. The whole pre-chat generation, in one notice.
- OpenAIdeprecationsource ↗
The first pinned GPT-4 and GPT-3.5 snapshots are deprecated
gpt-4-0314 is scheduled for 13 June 2024 and gpt-3.5-turbo-0301 for 13 September 2024, the first time a dated snapshot is given a year to live.
- OpenAIdeprecationsource ↗
The Codex models are deprecated with three days of notice
code-davinci-002, code-cushman-002 and their predecessors are scheduled for shutdown on 23 March. The shortest notice period in the record, by a factor of twenty.
- Anthropiclaunchsource ↗
Claude and Claude Instant open the Anthropic API
Two versions on the first day: a high-performance model, and a lighter, faster, cheaper one. The oldest event the record can primary-source.
An event appears here only with a primary-source URL verified on the date the entry carries. Secondary coverage can corroborate a date but never stands as the source. A detected price move is published only after a second read at least an hour later agrees, and cites the archived document either side of it.
A week with 3 events or more is a page of its own, with its counts, its sized moves and every entry in it. The newest is the week of 16 Nov 2026, with 16 price moves, and all 16 are listed by week.
This log is published as a JSON Feed at /api/v1/feed.json, public and with no key, carrying every entry here and the decisions the engine has proposed against them, and as RSS at /log/rss.xml for a feed reader, with the same items under the same ids. Every headline above is also a page of its own, at an address that keeps pointing at it. Nothing in either feed is a recommendation to act.
