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 · 601–650
| # | Model | Creator | Intelligence | $/1M in | $/1M out | tok/s |
|---|---|---|---|---|---|---|
| 601 | Phi-4 Multimodal Instruct | Microsoft | 5.80 | $0.08 | $0.32 | — |
| 602 | Mistral Large (Feb '24) | Mistral | 5.80 | $4 | $12 | — |
| 603 | Mistral Small (Sep '24) | Mistral | 5.80 | — | — | — |
| 604 | NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) | NVIDIA | 5.80 | $0.2 | $0.6 | 206 |
| 605 | Gemma 3n E4B Instruct Preview (May '25) | 5.80 | — | — | — | |
| 606 | Phi-3 Mini Instruct 3.8B | Microsoft | 5.80 | — | — | — |
| 607 | Qwen2.5 Coder Instruct 7B | Alibaba | 5.80 | — | — | — |
| 608 | MiniCPM-V 4.6 1.3B | OpenBMB | 5.70 | — | — | — |
| 609 | Llama 3.2 Instruct 3B | Meta | 5.70 | — | — | — |
| 610 | Mixtral 8x22B Instruct | Mistral | 5.70 | — | — | — |
| 611 | Qwen1.5 Chat 110B | Alibaba | 5.70 | — | — | — |
| 612 | Qwen3 VL 4B Instruct | Alibaba | 5.70 | — | — | — |
| 613 | Jamba Reasoning 3B | AI21 Labs | 5.70 | — | — | — |
| 614 | Llama 2 Chat 7B | Meta | 5.70 | $0.05 | $0.25 | — |
| 615 | Claude 3 Haiku | Anthropic | 5.60 | $0.25 | $1.25 | — |
| 616 | Olmo 3 7B Think | Allen Institute for AI | 5.60 | — | — | — |
| 617 | Claude 2.1 | Anthropic | 5.60 | — | — | — |
| 618 | Molmo 7B-D | Allen Institute for AI | 5.60 | — | — | — |
| 619 | Reka Flash 3 | Reka AI | 5.60 | $0.2 | $0.8 | 88 |
| 620 | OLMo 2 7B | Allen Institute for AI | 5.60 | — | — | — |
| 621 | GPT-3.5 Turbo | OpenAI | 5.50 | $0.5 | $1.5 | — |
| 622 | Mistral Medium | Mistral | 5.50 | — | — | — |
| 623 | DeepSeek-V2-Chat | DeepSeek | 5.50 | — | — | — |
| 624 | Mistral Small (Feb '24) | Mistral | 5.50 | — | — | — |
| 625 | Ling-mini-2.0 | InclusionAI | 5.50 | — | — | — |
| 626 | DeepSeek R1 Distill Qwen 1.5B | DeepSeek | 5.50 | — | — | — |
| 627 | Ministral 3 8B | Mistral | 5.50 | $0.15 | $0.15 | 105 |
| 628 | Llama 3 Instruct 70B | Meta | 5.50 | $0.65 | $2.75 | — |
| 629 | Claude 2.0 | Anthropic | 5.50 | — | — | — |
| 630 | LFM 40B | Liquid AI | 5.40 | — | — | — |
| 631 | Qwen3.5 0.8B (Non-reasoning) | Alibaba | 5.40 | — | — | — |
| 632 | Qwen Chat 72B | Alibaba | 5.40 | — | — | — |
| 633 | PALM-2 | 5.40 | — | — | — | |
| 634 | Llama 3.2 Instruct 11B (Vision) | Meta | 5.40 | $0.345 | $0.345 | 23 |
| 635 | Arctic Instruct | Snowflake | 5.40 | — | — | — |
| 636 | DBRX Instruct | Databricks | 5.30 | — | — | — |
| 637 | Sarvam M (Reasoning, Based on Mistral Small 3.1) | Sarvam | 5.30 | $0 | $0 | — |
| 638 | Llama 2 Chat 70B | Meta | 5.30 | — | — | — |
| 639 | Llama 2 Chat 13B | Meta | 5.30 | — | — | — |
| 640 | Command-R+ (Apr '24) | Cohere | 5.30 | — | — | — |
| 641 | DeepSeek LLM 67B Chat (V1) | DeepSeek | 5.30 | — | — | — |
| 642 | Gemini 1.0 Pro | 5.30 | — | — | — | |
| 643 | DeepSeek Coder V2 Lite Instruct | DeepSeek | 5.30 | — | — | — |
| 644 | Exaone 4.0 1.2B (Reasoning) | LG AI Research | 5.30 | — | — | — |
| 645 | OpenChat 3.5 (1210) | OpenChat | 5.30 | — | — | — |
| 646 | LFM2.5-1.2B-Instruct | Liquid AI | 5.20 | — | — | — |
| 647 | Jamba 1.5 Mini | AI21 Labs | 5.20 | $0.2 | $0.4 | — |
| 648 | Qwen3 1.7B (Reasoning) | Alibaba | 5.20 | — | — | — |
| 649 | LFM2.5-1.2B-Thinking | Liquid AI | 5.20 | — | — | — |
| 650 | LFM2 2.6B | Liquid AI | 5.20 | — | — | — |
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.