Open-Source Alternatives to 
GPT-3.5 Turbo (batch)
The best open-weight, self-hostable alternatives to GPT-3.5 Turbo (batch) in 2026 — compared on price, context window and capabilities. Run them locally and cut API costs.
Refreshed from live data · olud.ai
The best open-weight alternatives to GPT-3.5 Turbo (batch)
GPT-3.5 Turbo (batch) is a proprietary, API-only model. These open-weight models can be self-hosted, run offline and used at a fraction of the cost — here's how the top ones stack up.
DeepSeek V4.1 Flash Open
34.0 pts ABOVE GPT-3.5 Turbo (batch) on the Artificial Analysis intelligence index · 1.3× cheaper per million output tokens
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...
DeepSeek V4.1 Flash vs GPT-3.5 Turbo (batch) →
Hy3 Open
20.3 pts ABOVE GPT-3.5 Turbo (batch) on the Artificial Analysis intelligence index · 2.3× cheaper per million output tokens
Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...
Hy3 vs GPT-3.5 Turbo (batch) →IN Ling 3.0 Flash Open
19.4 pts ABOVE GPT-3.5 Turbo (batch) on the Artificial Analysis intelligence index · 12.5× cheaper per million output tokens
*Ling-3.0-flash* is a *124B-parameter Mixture-of-Experts (MoE) model*, with approximately *5.1B parameters activated per token*. The model is designed with *token efficiency and production-scale agentic inference* as key priorities, enablin
Ling 3.0 Flash vs GPT-3.5 Turbo (batch) →
Gemma 4 26B A4B Open
11.2 pts ABOVE GPT-3.5 Turbo (batch) on the Artificial Analysis intelligence index · 3.4× cheaper per million output tokens
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...
Gemma 4 26B A4B vs GPT-3.5 Turbo (batch) →
North Mini Code (free) Open
7.3 pts ABOVE GPT-3.5 Turbo (batch) on the Artificial Analysis intelligence index
North Mini Code is Cohere's first agentic coding model and the debut of its North family. A sparse mixture-of-experts model with 30B total parameters and 3B active, it is optimized...
North Mini Code (free) vs GPT-3.5 Turbo (batch) → gpt-oss-120b (batch) Open
6.8 pts ABOVE GPT-3.5 Turbo (batch) on the Artificial Analysis intelligence index · 1.3× cheaper per million output tokens
gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimize
gpt-oss-120b (batch) vs GPT-3.5 Turbo (batch) →See the full open-source leaderboard
Live ranking of open-weight models with pricing, context windows and capabilities.
Open the leaderboard →