Journal des modifications des modèles d'IA · 2026-07-16

117 événements enregistrés le 2026-07-16.

Modèle ajouté (4)

Moonshot AI New model: Kimi For Coding HighSpeed
amazon-bedrock New model: Qwen3 Coder 30B A3B Instruct
privatemode-ai New model: Voxtral Mini 3B
vercel New model: Qwen 3 Coder 30B A3B Instruct

Limite de sortie modifiée (43)

Alibaba (Qwen)Max output: Qwen 3.6 Plus updated
Alibaba (Qwen)Max output: Qwen2.5 VL 32B Instruct updated
Alibaba (Qwen)Max output: Qwen3 VL 235B A22B Instruct updated
Alibaba (Qwen)Max output: Qwq 32B updated
Anthropic Max output: anthropic--claude-3.5-sonnet updated
pioneer Max output: LFM2 24B A2B updated
Anthropic Max output: Anthropic: Claude Opus 4.5 updated
Anthropic Max output: Anthropic: Claude Sonnet 4.5 updated
baidu Max output: ERNIE 4.5 VL 28B A3B updated
DeepSeek Max output: DeepSeek-R1 updated
DeepSeek Max output: Deepseek R1 Distill Qwen 32B updated
DeepSeek Max output: DeepSeek-V3.2 updated
Google Max output: Gemma 4 31B updated
Meta Max output: Hermes 2 Pro Llama 3 8B updated
Meta Max output: Llama 3.1 70B Hanami updated
Meta Max output: Llama 3.3 70B updated
Meta Max output: DeepSeek R1 Distill Llama 70B updated
MiniMax Max output: MiniMax M1 80K updated
Mistral Max output: Mistral: Mistral 7B Instruct v0.1 updated
Mistral Max output: Mistral Medium 3.5 updated
Mistral Max output: Mistral Nemo updated
Mistral Max output: Mistral Small 3.2 updated
Mistral Max output: mistral-small-4-119b-2603 updated
nano-gpt Max output: GLM 5.1 Thinking TEE updated
nano-gpt Max output: Kimi K2.5 Thinking updated
nano-gpt Max output: MiniMax Latest updated
nano-gpt Max output: DeepSeek V4 Pro (Thinking) updated
OpenAI Max output: Whisper Large v3 updated
openrouter Max output: DeepSeek V3 0324 updated
openrouter Max output: Nex-N2-Pro (free) updated
pioneer Max output: DiffusionGemma 26B-A4B IT updated
pioneer Max output: Gemma 3 4B (Pretrained) updated
pioneer Max output: Gemma 4 12B IT updated
pioneer Max output: Gemma 4 E2B IT updated
pioneer Max output: Gemma 4 E4B IT updated
pioneer Max output: Qwen3 1.7B Base updated
pioneer Max output: Qwen3 4B Base updated
pioneer Max output: Qwen3 4B Instruct updated
pioneer Max output: SmolLM3 3B Base updated
xAI Max output: Grok-4 updated
xAI Max output: Grok 4 Fast updated
xAI Max output: Grok-4-Fast-Non-Reasoning updated
xAI Max output: Grok Code Fast 1 updated

Capacité mise à jour (41)

Alibaba (Qwen)Capabilities updated: Qwen 3.6 Plus
Alibaba (Qwen)Capabilities updated: Qwq 32B
Anthropic Capabilities updated: Claude Haiku 3.5
Anthropic Capabilities updated: anthropic--claude-3.5-sonnet
Anthropic Capabilities updated: Claude 3 Opus
Anthropic Capabilities updated: Anthropic: Claude 4.5 Haiku
Anthropic Capabilities updated: Anthropic: Claude Opus 4.5
Anthropic Capabilities updated: Anthropic: Claude Sonnet 4.5
baidu Capabilities updated: ERNIE 4.5 300B A47B
baidu Capabilities updated: ERNIE 4.5 VL 28B A3B
ByteDance (Doubao)Capabilities updated: Doubao-Seed-2.0-pro
DeepSeek Capabilities updated: DeepSeek-R1
DeepSeek Capabilities updated: Deepseek R1 Distill Qwen 32B
DeepSeek Capabilities updated: DeepSeek-V3.2
Google Capabilities updated: Gemini 2.5 Flash Lite Preview (09/2025)
Google Capabilities updated: Gemini Flash Latest
Google Capabilities updated: Gemini Flash-Lite Latest
Google Capabilities updated: Google Gemini Pro Latest
Google Capabilities updated: Gemma 4 31B
Meta Capabilities updated: Hermes 2 Pro Llama 3 8B
Meta Capabilities updated: Llama 3.1 70B Hanami
Meta Capabilities updated: Llama 3.3 70B
MiniMax Capabilities updated: MiniMax M1 80K
Mistral Capabilities updated: Mistral: Mistral 7B Instruct v0.1
Mistral Capabilities updated: Mistral-7B-Instruct-v0.3
Mistral Capabilities updated: Mistral Medium 3.5
Mistral Capabilities updated: Mistral Nemo
Mistral Capabilities updated: Mistral Small 3.2
Mistral Capabilities updated: mistral-small-4-119b-2603
nano-gpt Capabilities updated: GLM 5.1 Thinking TEE
nano-gpt Capabilities updated: Kimi K2.5 Thinking
nano-gpt Capabilities updated: MiniMax Latest
OpenAI Capabilities updated: ChatGPT-4o-Latest
openrouter Capabilities updated: OpenAI GPT Latest
openrouter Capabilities updated: Nex-N2-Pro (free)
xAI Capabilities updated: Grok 3 Beta
xAI Capabilities updated: Grok 3 Mini Beta
xAI Capabilities updated: Grok-4
xAI Capabilities updated: Grok 4 Fast
xAI Capabilities updated: Grok-4-Fast-Non-Reasoning
xAI Capabilities updated: Grok Code Fast 1

Contexte modifié (20)

Alibaba (Qwen)Context window: Qwen2.5 VL 32B Instruct now 16,384 (was 131,072)
Alibaba (Qwen)Context window: Qwen3 VL 235B A22B Instruct now 131,072 (was 262,144)
Alibaba (Qwen)Context window: Qwq 32B now 32,768 (was 24,000)
baidu Context window: ERNIE 4.5 VL 28B A3B now 32,768 (was 30,000)
DeepSeek Context window: DeepSeek-R1 now 128,000 (was 163,840)
DeepSeek Context window: Deepseek R1 Distill Qwen 32B now 80,000 (was 32,768)
DeepSeek Context window: DeepSeek V3.1 Terminus now 131,072 (was 163,840)
DeepSeek Context window: DeepSeek-V3.2 now 163,840 (was 128,000)
Google Context window: Gemini 2.5 Flash Lite Preview (09/2025) now 1,048,756 (was 1,000,000)
Google Context window: Gemma 4 31B now 262,144 (was 256,000)
Meta Context window: Hermes 2 Pro Llama 3 8B now 131,072 (was 8,192)
Meta Context window: Llama 3.1 70B Hanami now 16,384 (was 16,000)
Meta Context window: DeepSeek R1 Distill Llama 70B now 128,000 (was 32,768)
Mistral Context window: Mistral Nemo now 16,384 (was 128,000)
Mistral Context window: Mistral Small 3.2 24B now 131,072 (was 128,000)
Mistral Context window: Mistral Small 3.2 now 128,000 (was 32,000)
nano-gpt Context window: GLM 5.1 Thinking TEE now 202,752 (was 200,000)
nano-gpt Context window: Kimi K2.5 Thinking now 128,000 (was 256,000)
nano-gpt Context window: MiniMax Latest now 512,000 (was 1,048,576)
nano-gpt Context window: DeepSeek V4 Pro (Thinking) now 800,000 (was 1,048,576)

Prix modifié (9)

Alibaba (Qwen)Price update: Qwen3 30B A3B Instruct 2507 now $0.09/$0.3 (was $0.04815/$0.19305)
Google Price update: Gemini Flash Latest now $1.5/$9 (was $0.3/$2.5)
Google Price update: Gemini Flash-Lite Latest now $0.25/$1.5 (was $0.1/$0.4)
nano-gpt Price update: GLM 5.1 Thinking TEE now $1.5/$5.25 (was $0.3/$2.55)
nano-gpt Price update: MiniMax Latest now $0.3/$1.2 (was $1.1/$2.2)
nano-gpt Price update: DeepSeek V4 Pro (Thinking) now $1.5/$5.25 (was $1.1/$2.2)
OpenAI Price update: gpt-oss-120b now $0.04/$0.14 (was $0.03/$0.15)
openrouter Price update: DeepSeek V3 0324 now $0.27/$1.12 (was $0.24/$0.9)
openrouter Price update: Hy3 now $0.2/$0.8 (was $0.14/$0.58)

Pricing and capabilities are refreshed daily and reconciled against each provider's official documentation. Always verify critical production decisions with the provider directly.