blog••
6
大模型 API 生态日报 —— 2026-11-17
无人值守每日汇总大模型 API 生态:4 条发布动态、1 个新模型、34 项价格变动与 1 条社区热议。
一、今日速览
- 覆盖模型:470
- 新增模型:1
- 价格变动:34
- 社区信号:1
本篇日报由 AINode 无人值守内容引擎自动生成,信源包括公开模型库、GitHub 官方发布与开发者社区讨论。
二、新增模型
| 模型 ID | 厂商 | 上下文 | 输入 /1M | 输出 /1M |
|---|---|---|---|---|
z-ai/glm-5.2:batch |
z-ai | 1.0M | $0.322 | $0.989 |
三、价格变动
tencent/hy4-flash输入 $0.044 → $0.035 (−20%)tencent/hy4-flash输出 $0.176 → $0.140 (−20%)qwen/qwen4-max输入 $1.50 → $1.42 (−5%)qwen/qwen4-max输出 $4.50 → $4.27 (−5%)moonshotai/kimi-k3.1-pro:batch输入 $0.381 → $0.372 (−2%)moonshotai/kimi-k3.1-pro:batch输出 $5.71 → $5.58 (−2%)deepseek/deepseek-v4.1-flash-1017输入 $0.150 → $0.036 (−76%)deepseek/deepseek-v4.1-flash-1017输出 $0.600 → $0.500 (−17%)deepseek/deepseek-v4.1-flash-1017:batch输入 $0.037 → $0.036 (−3%)deepseek/deepseek-v4.1-flash-1017:batch输出 $0.326 → $0.150 (−54%)z-ai/glm-5.3-prime输入 $2.80 → $1.68 (−40%)z-ai/glm-5.3-prime输出 $8.80 → $5.28 (−40%)
四、官方与开源发布动态
- vllm-project/vllm v0.34.0:v0.34.0
- 10 new models (Mistral Large 4, MiniMax-M3.1-Flash, Nemotron-3.5-Transcribe, and more!)
- Extensible KV cache is on by default: vLLM sizes the KV cache from memory actually free after warmup, so most models get a bigger cache and more throughput.
- Model Runner V1 is gone. Model Runner V2 is now the only GPU runner.
- Big speedups on Blackwell and MI355X: up to 1.95× decode (Step5 on GB200), 1.6× throughput (GLM-5.3-Flash), 1.24–1.40× faster long-prompt TTFT (GLM-5.3).
- ollama/ollama v0.40.17:v0.40.17
- Improve ChatGPT CLI restore and graceful restart by @hoyyeva in #19583
- cmd/launch: use thinking metadata for OpenCode settings by @hoyyeva in #19409
- llm: use mmap for Vulkan integrated GPUs by @dhiltgen in #19640
- rpreslar4765/Ollama by @rpreslar4765 in #17163
- deepseek-ai/DeepSeek-V3 v1.0.0:v1.0.0 查看发布 ↗
- meta-llama/llama-models v0.2.0:v0.2.0 查看发布 ↗
五、社区热议
六、网关小贴士
统一 OpenAI 兼容入口与单一密钥,跨厂商做故障转移、Token 计量与按模型计费,无需逐个对接厂商 SDK。