blog••
2
AI model ecosystem briefing — 2026-11-18
Unattended daily digest of the LLM API landscape: 4 releases, 8 new models, 17 pricing updates and 1 community signal.
1. Today at a glance
- Models tracked: 478
- New models: 8
- Pricing updates: 17
- Community signals: 1
This briefing is generated by the unattended AINode content engine from public model catalogs, GitHub releases and developer community discussions.
2. New models
| Model ID | Provider | Context | Input /1M | Output /1M |
|---|---|---|---|---|
openai/gpt-6.2-astra-pro |
openai | 1.1M | $8.00 | $40.00 |
openai/gpt-6.2-astra-pro:batch |
openai | 1.1M | $4.00 | $20.00 |
openai/gpt-6.2-astra |
openai | 1.1M | $8.00 | $40.00 |
openai/gpt-6.2-astra:batch |
openai | 1.1M | $4.00 | $20.00 |
openai/gpt-6.2-sol-pro |
openai | 1.1M | $2.00 | $10.00 |
openai/gpt-6.2-sol-pro:batch |
openai | 1.1M | $1.00 | $5.00 |
openai/gpt-6.2-sol |
openai | 1.1M | $2.00 | $10.00 |
openai/gpt-6.2-sol:batch |
openai | 1.1M | $1.00 | $5.00 |
3. Pricing updates
~openai/gpt-astra-latestinput $10.00 → $8.00 (−20%)~openai/gpt-astra-latestoutput $50.00 → $40.00 (−20%)deepseek/deepseek-v4.1-flashinput $0.300 → $0.051 (−83%)deepseek/deepseek-v4.1-flashoutput $1.20 → $0.203 (−83%)qwen/qwen3.8-27binput $0.150 → $0.060 (−60%)qwen/qwen3.8-27boutput $1.88 → $2.20 (+17%)moonshotai/kimi-k2.7-codeinput $0.700 → $0.600 (−14%)xiaomi/mimo-v2.5-proinput $0.435 → $0.480 (+10%)xiaomi/mimo-v2.5-prooutput $0.870 → $1.80 (+107%)qwen/qwen3.5-27binput $0.260 → $0.300 (+15%)qwen/qwen3.5-27boutput $2.60 → $2.40 (−8%)qwen/qwen3-next-80b-a3b-instructinput $0.090 → $0.100 (+11%)
4. Official & open-source releases
- 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 View release ↗
- meta-llama/llama-models v0.2.0:v0.2.0 View release ↗
5. Community pulse
- Show HN: I "invented" Model-as-a-Service for 95% Predictable Private AI (2 points · 0 comments)
6. Gateway tip
Keep one OpenAI-compatible endpoint and one key across providers. Route failover, meter tokens and bill by model without touching provider SDKs.