blog••
15
AI model ecosystem briefing — 2026-11-07
Unattended daily digest of the LLM API landscape: 4 releases, 7 new models, 15 pricing updates and 1 community signal.
1. Today at a glance
- Models tracked: 460
- New models: 7
- Pricing updates: 15
- 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 |
|---|---|---|---|---|
liquid/lfm-3-1.7b |
liquid | 262K | $0.080 | $0.200 |
liquid/lfm-3-1.7b:free |
liquid | 262K | Free | Free |
qwen/qwen3.8-27b:batch |
qwen | 262K | $0.160 | $2.00 |
meta/muse-glimmer-30b:batch |
meta | 131K | $0.240 | $0.960 |
deepseek/deepseek-v4-flash-0731:batch |
deepseek | 1.0M | $0.048 | $0.144 |
thinkingmachines/inkling-small:batch |
thinkingmachines | 524K | $0.360 | $0.960 |
minimax/minimax-m3:batch |
minimax | 524K | $0.224 | $0.880 |
3. Pricing updates
inclusionai/ling-3.1-flashinput $0.075 → $0.045 (−40%)inclusionai/ling-3.1-flashoutput $0.225 → $0.135 (−40%)~z-ai/glm-flash-latestinput $0.028 → $0.029 (+0%)~deepseek/deepseek-v4-flash-latestinput $0.059 → $0.047 (−20%)deepseek/deepseek-v4-flash-0731input $0.059 → $0.047 (−20%)~moonshotai/kimi-latestinput $0.153 → $0.155 (+1%)~moonshotai/kimi-latestoutput $9.34 → $9.75 (+4%)deepseek/deepseek-v4-proinput $0.897 → $0.718 (−20%)deepseek/deepseek-v4-prooutput $1.79 → $1.44 (−20%)google/gemma-4-26b-a4b-itinput $0.090 → $0.068 (−25%)google/gemma-4-26b-a4b-itoutput $0.300 → $0.225 (−25%)minimax/minimax-m2.5input $0.270 → $0.300 (+11%)
4. Official & open-source releases
- ollama/ollama v0.40.15:v0.40.15
- ollama launch now supports Kimi Code CLI — the successor to kimi-cli — and offers to migrate existing kimi-cli setups
- ollama launch now fills in Pi's model list automatically so models can be picked inside Pi, and Claude Desktop's model picker shows Ollama's actual model names
- Recurrent and vision GGUF models can now serve more than one request at a time
- Qwen3 and Qwen3.5 models using long-context (YaRN) scaling
- vllm-project/vllm v0.33.0:v0.33.0 View release ↗
- 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.