LLM leaderboard
Every tracked language model by intelligence, with what it costs and how fast it answers.
ranked by intelligence · 611 models · updated Aug 30 · 501–550
| # | Model | Creator | Intelligence | $/1M in | $/1M out | tok/s |
|---|---|---|---|---|---|---|
| 501 | Llama 3.2 Instruct 90B (Vision) | Meta | 6.00 | — | — | — |
| 502 | Reka Flash (Sep '24) | Reka AI | 6.00 | $0.2 | $0.8 | 12 |
| 503 | Solar Mini | Upstage | 6.00 | $0.15 | $0.15 | — |
| 504 | Grok-1 | SpaceXAI | 5.80 | — | — | — |
| 505 | EXAONE 4.0 32B (Non-reasoning) | LG AI Research | 5.70 | — | — | — |
| 506 | Qwen2 Instruct 72B | Alibaba | 5.70 | — | — | — |
| 507 | Phi-4 Mini Instruct | Microsoft | 5.70 | $0 | $0 | 43 |
| 508 | Gemma 3 12B Instruct | 5.50 | $0 | $0 | — | |
| 509 | Qwen3.5 2B (Non-reasoning) | Alibaba | 5.30 | — | — | — |
| 510 | Gemini 1.5 Flash-8B | 5.20 | — | — | — | |
| 511 | Qwen3.5 0.8B (Reasoning) | Alibaba | 5.20 | — | — | — |
| 512 | Jamba 1.7 Large | AI21 Labs | 5.00 | — | — | — |
| 513 | DeepHermes 3 - Mistral 24B Preview (Non-reasoning) | Nous Research | 5.00 | — | — | — |
| 514 | Granite 4.0 H Small | IBM | 4.90 | $0.06 | $0.25 | 26 |
| 515 | Qwen3 Omni 30B A3B Instruct | Alibaba | 4.80 | $0.25 | $0.97 | 92 |
| 516 | Hermes 3 - Llama-3.1 70B | Nous Research | 4.80 | $0.7 | $0.7 | 34 |
| 517 | Qwen3 8B (Non-reasoning) | Alibaba | 4.80 | $0.18 | $0.7 | 39 |
| 518 | Jamba 1.5 Large | AI21 Labs | 4.80 | $2 | $8 | — |
| 519 | OLMo 2 32B | Allen Institute for AI | 4.70 | — | — | — |
| 520 | DeepSeek-Coder-V2 | DeepSeek | 4.70 | — | — | — |
| 521 | Jamba 1.6 Large | AI21 Labs | 4.70 | — | — | — |
| 522 | Gemini 1.5 Flash (May '24) | 4.60 | — | — | — | |
| 523 | Phi-4 | Microsoft | 4.60 | $0.13 | $0.5 | 41 |
| 524 | LFM2 24B A2B | Liquid AI | 4.60 | — | — | — |
| 525 | Nova Micro | Amazon | 4.40 | $0.04 | $0.14 | 264 |
| 526 | Granite 4.1 3B | IBM | 4.40 | — | — | — |
| 527 | Claude 3 Sonnet | Anthropic | 4.40 | — | — | — |
| 528 | Gemini 1.0 Ultra | 4.30 | — | — | — | |
| 529 | Mistral Small (Sep '24) | Mistral | 4.30 | $0.2 | $0.6 | 139 |
| 530 | Phi-3 Mini Instruct 3.8B | Microsoft | 4.30 | — | — | — |
| 531 | Phi-4 Multimodal Instruct | Microsoft | 4.20 | $0 | $0 | 18 |
| 532 | NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) | NVIDIA | 4.20 | $0.2 | $0.6 | 169 |
| 533 | Gemma 3n E4B Instruct Preview (May '25) | 4.20 | — | — | — | |
| 534 | Mistral Large (Feb '24) | Mistral | 4.10 | $4 | $12 | — |
| 535 | Qwen2.5 Coder Instruct 7B | Alibaba | 4.10 | — | — | — |
| 536 | Mixtral 8x22B Instruct | Mistral | 4.00 | — | — | — |
| 537 | Llama 3.2 Instruct 3B | Meta | 3.90 | — | — | — |
| 538 | Llama 2 Chat 7B | Meta | 3.90 | $0.05 | $0.25 | — |
| 539 | MiniCPM-V 4.6 1.3B | OpenBMB | 3.80 | — | — | — |
| 540 | Jamba Reasoning 3B | AI21 Labs | 3.80 | — | — | — |
| 541 | Qwen1.5 Chat 110B | Alibaba | 3.70 | — | — | — |
| 542 | Reka Flash 3 | Reka AI | 3.70 | $0.2 | $0.8 | — |
| 543 | Qwen3 VL 4B Instruct | Alibaba | 3.70 | — | — | — |
| 544 | Olmo 3 7B Think | Allen Institute for AI | 3.60 | — | — | — |
| 545 | Claude 3 Haiku | Anthropic | 3.50 | $0.25 | $1.25 | — |
| 546 | Claude 2.1 | Anthropic | 3.50 | — | — | — |
| 547 | OLMo 2 7B | Allen Institute for AI | 3.50 | — | — | — |
| 548 | Ling-mini-2.0 | InclusionAI | 3.40 | — | — | — |
| 549 | Molmo 7B-D | Allen Institute for AI | 3.40 | — | — | — |
| 550 | DeepSeek-V2-Chat | DeepSeek | 3.30 | — | — | — |
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.