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 · 451–500
| # | Model | Creator | Intelligence | $/1M in | $/1M out | tok/s |
|---|---|---|---|---|---|---|
| 451 | Devstral Small (May '25) | Mistral | 8.70 | — | — | — |
| 452 | MiniCPM5-1B (Non-reasoning) | OpenBMB | 8.70 | — | — | — |
| 453 | Devstral 2 | Mistral | 8.60 | — | — | — |
| 454 | Magistral Small 1.2 | Mistral | 8.60 | $0.5 | $1.5 | 74 |
| 455 | DeepSeek V3 (Dec '24) | DeepSeek | 8.50 | $0.32 | $0.89 | 37 |
| 456 | Gemini 2.5 Flash-Lite (Reasoning) | 8.50 | $0.1 | $0.4 | 331 | |
| 457 | GLM-4.6V (Non-reasoning) | Z AI | 8.40 | $0.3 | $0.9 | 82 |
| 458 | Qwen3 VL 32B Instruct | Alibaba | 8.40 | — | — | — |
| 459 | Nanbeige4.1-3B | Nanbeige | 8.40 | — | — | — |
| 460 | GPT-4o (Nov '24) | OpenAI | 8.40 | $2.5 | $10 | 160 |
| 461 | DeepSeek R1 Distill Qwen 32B | DeepSeek | 8.40 | — | — | — |
| 462 | LFM2.5-2.6B | Liquid AI | 8.40 | — | — | — |
| 463 | Qwen3 235B A22B (Non-reasoning) | Alibaba | 8.30 | — | — | — |
| 464 | Magistral Small 1 | Mistral | 8.20 | — | — | — |
| 465 | Nova 2.0 Omni (Non-reasoning) | Amazon | 8.20 | $0.3 | $2.5 | — |
| 466 | EXAONE 4.0 32B (Reasoning) | LG AI Research | 8.20 | — | — | — |
| 467 | Qwen3 VL 8B (Reasoning) | Alibaba | 8.20 | — | — | — |
| 468 | Gemini 2.0 Flash (experimental) | 8.20 | $0 | $0 | — | |
| 469 | DeepSeek R1 0528 Qwen3 8B | DeepSeek | 8.10 | — | — | — |
| 470 | Qwen2.5 Max | Alibaba | 8.00 | — | — | — |
| 471 | Claude 3.5 Sonnet (Oct '24) | Anthropic | 7.90 | $3 | $15 | — |
| 472 | Gemini 1.5 Pro (Sep '24) | 7.90 | — | — | — | |
| 473 | Solar Pro 2 (Preview) (Non-reasoning) | Upstage | 7.90 | — | — | — |
| 474 | DeepSeek R1 Distill Llama 70B | DeepSeek | 7.90 | $0.7 | $1.1 | — |
| 475 | Qwen3 VL 30B A3B Instruct | Alibaba | 7.90 | $0.35 | $0.6 | 24 |
| 476 | Hermes 4 - Llama-3.1 70B (Reasoning) | Nous Research | 7.90 | — | — | — |
| 477 | GPT-4.1 nano | OpenAI | 7.80 | $0.1 | $0.4 | 212 |
| 478 | Ling-flash-2.0 | InclusionAI | 7.80 | $0.14 | $0.57 | 5 |
| 479 | Gemma 4 E2B (Reasoning) | 7.80 | — | — | — | |
| 480 | DeepSeek R1 Distill Qwen 14B | DeepSeek | 7.80 | — | — | — |
| 481 | Solar Pro 3 | Upstage | 7.80 | $0.15 | $0.6 | 159 |
| 482 | Qwen3 Omni 30B A3B (Reasoning) | Alibaba | 7.80 | $0.25 | $0.97 | 102 |
| 483 | Falcon-H1R-7B | TII UAE | 7.80 | — | — | — |
| 484 | Sonar | Perplexity | 7.70 | — | — | — |
| 485 | GPT-4o (Aug '24) | OpenAI | 7.70 | $2.5 | $10 | 147 |
| 486 | Llama 3.3 Instruct 70B | Meta | 7.70 | $0.71 | $0.72 | 93 |
| 487 | Step3 VL 10B | StepFun | 7.70 | — | — | — |
| 488 | Qwen2.5 Instruct 72B | Alibaba | 7.70 | $0.475 | $0.495 | 30 |
| 489 | GLM-4.5V (Reasoning) | Z AI | 7.60 | $0.6 | $1.8 | 47 |
| 490 | Sonar Pro | Perplexity | 7.60 | — | — | — |
| 491 | Qwen3 30B A3B (Reasoning) | Alibaba | 7.60 | $0.135 | $0.55 | 113 |
| 492 | Mistral Large 2 (Nov '24) | Mistral | 7.60 | — | — | — |
| 493 | QwQ 32B-Preview | Alibaba | 7.60 | — | — | — |
| 494 | Devstral Small (Jul '25) | Mistral | 7.60 | — | — | — |
| 495 | Gemma 4 E4B (Non-reasoning) | 7.50 | $0.02 | $0.1 | 35 | |
| 496 | NVIDIA Nemotron Nano 12B v2 VL (Reasoning) | NVIDIA | 7.50 | — | — | — |
| 497 | Qwen3 30B A3B 2507 Instruct | Alibaba | 7.50 | $0.2 | $0.4 | 44 |
| 498 | Hermes 4 - Llama-3.1 405B (Reasoning) | Nous Research | 7.50 | $1 | $3 | 40 |
| 499 | ERNIE 4.5 300B A47B | Baidu | 7.50 | $0.28 | $1.1 | — |
| 500 | Devstral Small 2 | Mistral | 7.50 | — | — | — |
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.