LLM leaderboard
Every tracked language model by intelligence, with what it costs and how fast it answers.
ranked by intelligence · 690 models · updated Oct 15 · 651–690
| # | Model | Creator | Intelligence | $/1M in | $/1M out | tok/s |
|---|---|---|---|---|---|---|
| 651 | Exaone 4.0 1.2B (Non-reasoning) | LG AI Research | 5.20 | — | — | — |
| 652 | Olmo 3 7B Instruct | Allen Institute for AI | 5.20 | $0.1 | $0.2 | — |
| 653 | Qwen3 8B (Reasoning) | Alibaba | 5.20 | — | — | — |
| 654 | Jamba 1.7 Mini | AI21 Labs | 5.20 | — | — | — |
| 655 | Granite 4.0 H 1B | IBM | 5.20 | — | — | — |
| 656 | Jamba 1.6 Mini | AI21 Labs | 5.20 | — | — | — |
| 657 | Gemma 3 270M | 5.10 | — | — | — | |
| 658 | Granite 4.0 Micro | IBM | 5.10 | — | — | — |
| 659 | Mixtral 8x7B Instruct | Mistral | 5.10 | $0.45 | $0.7 | 25 |
| 660 | Apertus 70B Instruct | Swiss AI Initiative | 5.10 | $0.82 | $2.92 | — |
| 661 | DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) | Nous Research | 5.10 | — | — | — |
| 662 | Mistral 7B Instruct | Mistral | 5.00 | $0.15 | $0.2 | — |
| 663 | Molmo2-8B | Allen Institute for AI | 5.00 | — | — | — |
| 664 | Claude Instant | Anthropic | 5.00 | — | — | — |
| 665 | Granite 4.0 1B | IBM | 5.00 | — | — | — |
| 666 | Command-R (Mar '24) | Cohere | 5.00 | — | — | — |
| 667 | Qwen Chat 14B | Alibaba | 5.00 | — | — | — |
| 668 | Llama 65B | Meta | 5.00 | — | — | — |
| 669 | Qwen3 1.7B (Non-reasoning) | Alibaba | 4.90 | — | — | — |
| 670 | Gemma 3 27B Instruct | 4.90 | $0 | $0 | — | |
| 671 | Granite 3.3 8B (Non-reasoning) | IBM | 4.90 | $0.03 | $0.25 | 44 |
| 672 | LFM2 8B A1B | Liquid AI | 4.90 | — | — | — |
| 673 | Qwen3 0.6B (Non-reasoning) | Alibaba | 4.80 | — | — | — |
| 674 | LFM2 1.2B | Liquid AI | 4.80 | — | — | — |
| 675 | Gemma 3 4B Instruct | 4.80 | $0 | $0 | — | |
| 676 | LFM2.5-VL-1.6B | Liquid AI | 4.80 | — | — | — |
| 677 | Qwen3 0.6B (Reasoning) | Alibaba | 4.80 | — | — | — |
| 678 | Granite 4.0 350M | IBM | 4.80 | — | — | — |
| 679 | Ministral 3 3B | Mistral | 4.80 | $0.1 | $0.1 | 218 |
| 680 | Llama 3.2 Instruct 1B | Meta | 4.80 | — | — | — |
| 681 | Tiny Aya Global | Cohere | 4.80 | $0 | $0 | — |
| 682 | Llama 3 Instruct 8B | Meta | 4.80 | $0.045 | $0.145 | — |
| 683 | Apertus 8B Instruct | Swiss AI Initiative | 4.80 | $0.1 | $0.2 | — |
| 684 | NVIDIA Nemotron 3 Nano 4B | NVIDIA | 4.80 | — | — | — |
| 685 | Granite 4.0 H 350M | IBM | 4.80 | — | — | — |
| 686 | Gemma 3n E4B Instruct | 4.80 | — | — | — | |
| 687 | Gemma 3n E2B Instruct | 4.80 | $0 | $0 | — | |
| 688 | Gemma 3 1B Instruct | 4.80 | $0 | $0 | — | |
| 689 | Gemma 3 12B Instruct | 3.80 | $0 | $0 | — | |
| 690 | K2 Horizon 0.9B | Institute of Foundation Models | 3.00 | — | — | — |
what intelligence means
A composite of nine independent evaluations covering reasoning, coding, agentic work and knowledge. Higher is better. It is versioned, so scores are comparable within a version rather than across all time.