2026-08-19

2026-08-19에 감지된 AI 모델 가격·컨텍스트·종료일 변경 전체입니다.

변경모델내용
종료 예정inclusionai/ling-2.6-1t2026-08-24 종료 · D-5 남음
종료 예정inclusionai/ling-2.6-flash2026-08-24 종료 · D-5 남음
종료 예정inclusionai/ring-2.6-1t2026-08-24 종료 · D-5 남음
종료 예정nvidia/nemotron-nano-12b-v2-vl:free2026-08-24 종료 · D-5 남음
종료 예정z-ai/glm-4.52026-12-31 종료 · D-134 남음
종료일 공지inclusionai/ling-2.6-1t- -> 2026-08-24
종료일 공지inclusionai/ling-2.6-flash- -> 2026-08-24
종료일 공지inclusionai/ring-2.6-1t- -> 2026-08-24
가격 변경deepseek/deepseek-chat-v3-0324출력 -10.7% · 캐시 읽기 과금 중단 · 입력 -7.4%
가격 변경deepseek/deepseek-v4-flash출력 +3.5% · 캐시 읽기 +3.5% · 입력 +3.5%
가격 변경deepseek/deepseek-v4-pro출력 +45.5% · 캐시 읽기 +452.3% · overrides 과금 중단 · 입력 +118.2%
가격 변경google/gemma-4-31b-it캐시 읽기 -50.0% · 입력 -10.0%
가격 변경minimax/minimax-m2.5캐시 읽기 +20.0% · 입력 +2.3%
가격 변경minimax/minimax-m3:batch출력 +100.0% · 캐시 읽기 +100.0% · 입력 +100.0%
가격 변경moonshotai/kimi-k2.5출력 -21.1% · 캐시 읽기 -26.3% · 입력 -21.1%
가격 변경moonshotai/kimi-k2.6출력 -3.4% · 캐시 읽기 -3.4% · 입력 -3.4%
가격 변경moonshotai/kimi-k2.7-code:batch출력 +100.0% · 캐시 읽기 +100.0% · 입력 +100.0%
가격 변경nvidia/nemotron-3-ultra-550b-a55b:batch출력 +100.0% · 캐시 읽기 +100.0% · 입력 +100.0%
가격 변경qwen/qwen3-next-80b-a3b-instruct캐시 읽기 과금 중단 · 입력 -10.0%
가격 변경qwen/qwen3.5-122b-a10b출력 -13.3% · 입력 -10.3%
가격 변경qwen/qwen3.5-35b-a3b출력 -30.6% · 캐시 읽기 +11.1% · 입력 +11.1%
가격 변경qwen/qwen3.6-27b출력 -16.7% · 캐시 읽기 과금 시작 · 입력 +3.8%
가격 변경thinkingmachines/inkling:batch출력 +100.0% · 캐시 읽기 +100.0% · 입력 +100.0%
가격 변경z-ai/glm-5.2출력 +102.9% · 캐시 읽기 +118.5% · 입력 +102.9%
가격 변경z-ai/glm-5.2:batch출력 +100.0% · 캐시 읽기 +100.0% · 입력 +100.0%
가격 변경~deepseek/deepseek-v4-flash-latest출력 -2.7% · 캐시 읽기 -2.7% · 입력 -2.7%
신규 모델z-ai/glm-5.3컨텍스트 1,048,576
신규 모델~z-ai/glm-latest컨텍스트 1,048,576
모델 삭제ai21/jamba-large-1.7
당사자 발언Sam Altman @sama(We still expect to ship great new models soon; this impacts further-out releases.)
당사자 발언Tibo @thsottiauxIt has not been used yet, but would you look at that. Codex for scale. https://t.co/o1pulwoifd
당사자 발언Tibo @thsottiaux@ClaudeDevs Do not worry, we have compute
당사자 발언Anthropic @AnthropicAIFor more on how Claude ran this experiment and the full results, see our blog: https://t.co/48COrpUcJn
당사자 발언Anthropic @AnthropicAIOne of our highest priorities remains launching an access program for scientists to use our most capable models. We expect to share more on this soon. Opus 5 remains our most capable model available for life science research.
당사자 발언Anthropic @AnthropicAIDesigning a binder is an easier process than designing a drug, but it’s a useful proxy. The typical success rate in the field today is between 10% and 15%. Between 22% and 35% of Claude's designs bound successfully, depending on the setup.
당사자 발언Google AI Developers @googleaidevsFrom a single prompt to a fully animated parallax landing page. 🤯 See how Gemini 3.7 Flash, combined with Nano Banana and Omni, builds interactive websites in one shot. Under the hood, 3.7 Flash calls the right tools to generate the copy, i
OUTAGEDegraded performance for Claude Opus 5 and Claude Haiku 4.5
RELEASEv0.124.0
RELEASEv2.1.235
RELEASE0.148.0
RELEASElangchain-core==1.6.0
RELEASEb10502
RELEASEv3.3.1
RELEASEPatch release: v5.15.1
ANNOUNCEMENTCredential revocation and deauthorization by token type
ANNOUNCEMENTEnterprise managed settings in GitHub Copilot for JetBrains
ANNOUNCEMENTTrack organization code quality trends
ANNOUNCEMENTGitHub Copilot app for Beginners: Managing your work
ANNOUNCEMENTLFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation
ANNOUNCEMENTChatGPT Ads expands across Europe
ANNOUNCEMENTHow NVIDIA scales expertise with ChatGPT Work
ANNOUNCEMENTOffering Zero Data Retention for frontier models
ANNOUNCEMENTReplit expands access to software creation with GPT-5.6 Luna
ANNOUNCEMENTStrengthening democratic oversight in national security
OUTAGEElevated errors deploying Sites
파라미터 변경deepseek/deepseek-chat-v3-0324추가 logprobs, top_logprobs
파라미터 변경inclusionai/ling-3.0-flash삭제 structured_outputs
파라미터 변경minimax/minimax-m2.5삭제 parallel_tool_calls

이전 2026-08-18