Open-Weight Models Grew Up
In 2026 the open-weight frontier is genuinely competitive — and dramatically cheaper than closed flagships. Four labs lead: DeepSeek, Z.ai (GLM), Moonshot (Kimi), and MiniMax. Each ships open weights and a hosted API, so you can call them like any API today and self-host later if you want. All four are on LinkModel under one key.
The Contenders
| Model | Input / Output (1M) | License | Edge |
|---|---|---|---|
| DeepSeek V4 Flash | $0.14 / $0.28 | MIT | Cheapest, 1M context, 98% cache |
| DeepSeek V4 Pro | ~$0.44 / $0.87 | MIT | Competition-grade coding |
| GLM-5.1 | ~$1.40 / $4.40 | MIT | 8-hour autonomous coding |
| Kimi K2.6 | $0.95 / $4.00 | Modified-MIT | Multimodal, 300 sub-agents |
| MiniMax M2.7 | $0.30 / $1.20 | Modified-MIT | Cheap long-horizon agents |
Which to Pick
- Cheapest / high-volume → DeepSeek V4 Flash. Nothing capable is cheaper.
- Best value coding → DeepSeek V4 Pro (LiveCodeBench 93.5).
- Long autonomous coding runs → GLM-5.1 (3rd on Code Arena, 8-hr stamina).
- Multimodal / UI-from-screenshot / agent swarms → Kimi K2.6.
- Cheapest capable agent → MiniMax M2.7.
Two Things to Know Before Self-Hosting
Licensing. DeepSeek and GLM-5.1 are fully MIT (commercial self-hosting OK). Kimi K2.6 and MiniMax M2.7 are Modified-MIT — commercial self-hosting needs authorization. Check the license against your deployment.
Data residency. These labs' official APIs process data on servers in China — a GDPR/compliance consideration for EU/PII workloads. Two clean options: self-host the open weights (your infra, your rules), or use a gateway with its own data handling. LinkModel defaults to zero data retention and runs all four behind one key, up to 30% below official — sidestepping both the license-for-hosting question and the residency issue.
API-First or Self-Host?
Open weights are "free" only if you own and operate the GPUs. For bursty or moderate volume, the hosted API is cheaper in total cost (no idle GPU, no ops) — see serverless GPU vs generation API. Self-host when you need data sovereignty, custom fine-tunes, or very high steady utilization.
How to Call Them
All four are OpenAI-compatible, so switching is a model-string change:
for model in ("deepseek-v4-flash", "glm-5.1", "kimi-k2.6", "minimax-m2.7"):
run(model, prompt) # same request shape via one gateway keyConfirm the exact chat endpoint in the docs.
Bottom Line
Open-weight LLMs are the value story of 2026: DeepSeek for cheap bulk, GLM for autonomous coding, Kimi for multimodal agents, MiniMax for cheap long-horizon agents. Mind the license and data residency, and route between them. Compare with closed flagships in best LLM API.
Start free with a $1 credit and test all four.