Best Models
Gemma 3n E4B Instruct is currently the highest-ranked model for Blended Price in the available data.
Use the top-ranked model as a starting point, not an automatic decision.
Compare the top choices on price, speed, availability, and your own constraints.
Open the linked result page and methodology before quoting the ranking.
Estimated cost per 1M tokens using Artificial Analysis's 3:1 input-to-output token mix. It blends input and output token prices into one comparison number; lower is cheaper.
| Rank | Model | Provider | Value |
|---|---|---|---|
| #1 | Gemma 3n E4B Instruct | $0.025 / 1M | |
| #2 | Sarvam 30B (high) | Sarvam | $0.047 / 1M |
| #3 | Qwen3.5 4B (Non-reasoning) | Alibaba | $0.06 / 1M |
| #4 | Qwen3.5 4B (Reasoning) | Alibaba | $0.06 / 1M |
| #5 | Nova Micro | Amazon |
This answer uses the exact same metric and live database ranking as the linked Easy Benchmarks leaderboard.
Source: Artificial Analysis Blended Price. Updated: Jul 16, 2026.
| $0.061 / 1M |
| #6 | Granite 4.1 8B | IBM | $0.063 / 1M |
| #7 | HyperNova 60B 2605 | Multiverse Computing | $0.065 / 1M |
| #8 | NVIDIA Nemotron Nano 9B V2 (Reasoning) | NVIDIA | $0.07 / 1M |
| #9 | Llama 3 Instruct 8B | Meta | $0.07 / 1M |
| #10 | Sarvam 105B (high) | Sarvam | $0.074 / 1M |
| #11 | Granite 3.3 8B (Non-reasoning) | IBM | $0.085 / 1M |
| #12 | NVIDIA Nemotron Nano 9B V2 (Non-reasoning) | NVIDIA | $0.086 / 1M |
| #13 | gpt-oss-20b (high) | OpenAI | $0.088 / 1M |
| #14 | NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) | NVIDIA | $0.088 / 1M |
| #15 | NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) | NVIDIA | $0.088 / 1M |
| #16 | Qwen2.5 Turbo | Alibaba | $0.088 / 1M |
| #17 | gpt-oss-20b (low) | OpenAI | $0.095 / 1M |
| #18 | Ministral 3 3B | Mistral | $0.10 / 1M |
| #19 | Llama 3.1 Instruct 8B | Meta | $0.10 / 1M |
| #20 | Llama 3.2 Instruct 1B | Meta | $0.10 / 1M |