所有LLM目录
综合性能最佳 LLM 排行榜
- 各任务收录模型
- 913
- 发布方
- 95
- 可算显存
- 370
- 01Hy4 flashTencent
- 02DeepSeek V4.1 Flash 1017DeepSeek
- 03DeepSeek V4.1 FlashDeepSeek
- 04GLM 5.3 FlashZ.ai
- 05Hy4Tencent
按 OpenRouter 一周处理的 token 数,在本目录的开放文本输出模型中排序,统计截至 · 数据来源 ↗
模型列表
有综合评分的模型 65
按 Artificial Analysis 综合能力指数排序。同分并列,名次以本目录收录的模型为范围。 评测方法 ↗
| 大模型 | 权重参数量 | 上下文长度 | 模态 | 发布日期 | 综合评分 | API 价格 / 1M |
|---|---|---|---|---|---|---|
| #1Claude Fable 5.2 | — | 1M | 2026-11-09 | 64.6 | ||
| #2GPT-6.2 Astra | — | 1.05M | 2026-11-17 | 62.1 | ||
| #3Gemini 4 Neon | — | 1.05M | — | 61.4 | ||
| #4Gemini 4 Carbon | — | 1.05M | 2026-11-10 | 60.0 | ||
| #5Claude Opus 5.5 | — | 1M | — | 59.1 | ||
| #6GPT-6.2 Sol | — | 1.05M | — | 57.3 | ||
| #7GPT-6 Astra | — | 1.05M | 2026-09-03 | 56.9 | ||
| #8GPT-6.1 Sol | — | 1.05M | 2026-09-29 | 54.6 | ||
| #9Claude Sonnet 5.5 | — | 1M | 2026-09-28 | 54.5 | ||
| #10Gemini 4 Hydrogen | — | 1.05M | 2026-11-12 | 53.9 | ||
| #11GPT-6.1 Terra | — | 1.05M | 2026-10-26 | 51.3 | ||
| #12Gemini 4 Argon | — | 1.05M | 2026-10-13 | 50.4 | ||
| #13Claude Fable 5.1 | — | 1M | — | 50.0 | ||
| #14Kimi K3.1 Pro | 2.78T | 1.05M | 2026-10-23 | 46.5 | ||
| #15Claude Opus 5 | — | 1M | — | 44.2 | ||
| #15GPT-6 Sol | — | 1.05M | 2026-09-22 | 44.2 | ||
| #17Muse Wonder | — | 1.05M | 2026-10-12 | 43.6 | ||
| #18Qwen4 Max | — | 1M | — | 42.9 | ||
| #19Muse Spark 1.3 | — | 1.05M | 2026-09-02 | 36.6 | ||
| #20Claude Haiku 5.5 | — | 1M | 2026-10-07 | 36.5 |
按模型估算显存 (370)
- Qwen3.5 27B
- Qwen3 8B
- Qwen3 14B
- Qwen3 32B
- Qwen2.5 7B Instruct
- Qwen2.5 32B Instruct
- Qwen2.5 72B Instruct
- Llama 3.1 8B Instruct
- Llama 3.2 3B Instruct
- Llama 3.3 70B Instruct
- DeepSeek-R1-Distill-Qwen 14B
- DeepSeek-R1-Distill-Qwen 32B
- DeepSeek-R1-Distill-Llama 8B
- Gemma 3 12B IT
- Gemma 3 27B IT
- Mistral 7B Instruct v0.3
- Mistral Small 24B Instruct 2501
- gpt-oss-20b
- gpt-oss-120b
- DeepSeek-R1 671B
- DeepSeek-R1 3B (suayptalha)
- deepseek-coder-1.3b-instruct
- DeepSeek-R1-Distill-Qwen-1.5B
- deepseek-coder-6.7b-base
- deepseek-coder-6.7b-instruct
- deepseek-coder-7b-instruct-v1.5
- deepseek-math-7b-rl
- DeepSeek-R1-Distill-Qwen-7B
- DeepSeek-R1-0528-Qwen3-8B
- DeepSeek-Coder-V2-Lite-Base
- DeepSeek-Coder-V2-Lite-Instruct
- DeepSeek-V2-Lite
- DeepSeek-V2-Lite-Chat
- deepseek-coder-33b-base
- deepseek-coder-33b-instruct
- DeepSeek-R1-Distill-Llama-70B
- DeepSeek-Coder-V2-Base
- DeepSeek-Coder-V2-Instruct
- DeepSeek-Coder-V2-Instruct-0724
- DeepSeek-V2
- DeepSeek-V2-Chat
- DeepSeek-V2-Chat-0628
- DeepSeek-V2.5
- DeepSeek-V2.5-1210
- Falcon3-1B-Base
- Falcon3-1B-Instruct
- Falcon3-3B-Base
- Falcon3-3B-Instruct
- Falcon3-7B-Base
- Falcon3-7B-Instruct
- Falcon3-10B-Base
- Falcon3-10B-Instruct
- granite-3.0-1b-a400m-instruct
- granite-3.1-1b-a400m-instruct
- granite-3.0-1b-a400m-base
- granite-3.1-2b-instruct
- granite-3.2-2b-instruct
- granite-3.3-2b-instruct
- granite-3.1-3b-a800m-base
- granite-3.1-3b-a800m-instruct
- granite-4.1-3b
- granite-4.1-3b-base
- granite-3b-code-base-2k
- granite-4.2-3b
- granite-3.3-8b-base
- granite-3.0-8b-instruct
- granite-3.1-8b-instruct
- granite-3.2-8b-instruct
- granite-3.3-8b-instruct
- granite-guardian-3.3-8b
- granite-4.1-8b-base
- granite-guardian-4.1-8b
- granite-4.1-8b
- granite-4.2-8b
- granite-4.1-30b
- granite-4.2-30b
- Llama-3.2-1B
- Hermes-3-Llama-3.2-3B
- Llama-2-7b-chat-hf
- Llama-2-7b-hf
- Hermes-2-Theta-Llama-3-8B
- Hermes-3-Llama-3.1-8B
- Meta-Llama-3-8B
- Meta-Llama-3-8B-Instruct
- Meta-Llama-3.1-8B
- Nous-Hermes-2-SOLAR-10.7B
- Llama-2-13b-hf
- Nous-Hermes-2-Yi-34B
- Nous-Hermes-2-Mixtral-8x7B-DPO
- Hermes-4-70B
- Hermes-4-405B
- Mistral-7B-Instruct-v0.1
- Mistral-7B-Instruct-v0.2
- Mistral-7B-v0.1
- OLMo-1B-hf
- OLMo-1B-0724-hf
- OLMo-2-0425-1B
- OLMo-2-0425-1B-Instruct
- OLMo-2-0425-1B-RLVR1
- OLMo-7B-hf
- Olmo-3-1025-7B
- Olmo-3-7B-Instruct
- Olmo-3-7B-Instruct-DPO
- Olmo-3-7B-Instruct-SFT
- Olmo-3-7B-RL-Zero-Math
- Olmo-3-7B-Think
- Olmo-3-7B-Think-DPO
- Olmo-3-7B-Think-SFT
- OLMo-2-1124-7B-Instruct
- Llama-3.1-Tulu-3-8B
- Llama-3.1-Tulu-3-8B-DPO
- Llama-3.1-Tulu-3-8B-SFT
- Llama-3.1-Tulu-3.1-8B
- OLMo-2-1124-13B-Instruct
- Olmo-3-1125-32B
- Olmo-3-32B-Think
- Olmo-3-32B-Think-SFT
- Olmo-3.1-32B-Instruct
- Olmo-3.1-32B-Instruct-DPO
- Olmo-3.1-32B-Think
- OLMo-2-0325-32B
- OLMo-2-0325-32B-Instruct
- phi-1
- phi-1_5
- phi-2
- phi-2-pytdml
- Phi-3-mini-128k-instruct
- Phi-3-mini-4k-instruct
- Phi-3.5-mini-instruct
- Phi-4-mini-instruct
- Phi-4-mini-reasoning
- Phi-3-medium-128k-instruct
- Phi-3-medium-4k-instruct
- Phi-4-reasoning
- Phi-4-reasoning-plus
- phi-4
- pythia-14m
- pythia-31m
- pythia-410m-addition-first-ft
- pythia-410m-capitals-first-ft
- pythia-410m-hemisphere-first-ft
- pythia-410m-modularaddition-first-ft
- pythia-410m-multiplication-first-ft
- pythia-410m-population-first-ft
- pythia-410m-sciq-first-ft
- pythia-410m-sentiment-first-ft
- pythia-410m-squaring-first-ft
- pythia-1.4b-deduped
- Qwen2-0.5B
- Qwen2-0.5B-Instruct
- Qwen2.5-0.5B
- Qwen2.5-0.5B-Instruct
- Qwen2.5-Coder-0.5B
- Qwen2.5-Coder-0.5B-Instruct
- Qwen3-0.6B-Base
- Qwen1.5-0.5B
- Qwen1.5-0.5B-Chat
- Qwen3-0.6B
- Qwen3Guard-Gen-0.6B
- Qwen2-1.5B
- Qwen2-1.5B-Instruct
- Qwen2.5-1.5B
- Qwen2.5-1.5B-Instruct
- Qwen2.5-Coder-1.5B
- Qwen2.5-Coder-1.5B-Instruct
- Qwen2.5-Math-1.5B
- Qwen2.5-Math-1.5B-Instruct
- Qwen3-1.7B-Base
- Qwen3-1.7B-MLX-bf16
- Qwen1.5-1.8B
- Qwen1.5-1.8B-Chat
- Qwen3-1.7B
- Qwen2.5-3B
- Qwen2.5-3B-Instruct
- Qwen2.5-Coder-3B
- Qwen2.5-Coder-3B-Instruct
- Qwen1.5-4B
- Qwen1.5-4B-Chat
- Qwen3-4B
- Qwen3-4B-Base
- Qwen3-4B-Instruct-2507
- Qwen3-4B-Thinking-2507
- Qwen3Guard-Gen-4B
- Qwen2-7B
- Qwen2-7B-Instruct
- Qwen2.5-7B
- Qwen2.5-7B-Instruct-1M
- Qwen2.5-Coder-7B
- Qwen2.5-Coder-7B-Instruct
- Qwen2.5-Math-7B
- Qwen2.5-Math-7B-Instruct
- Qwen1.5-7B
- Qwen1.5-7B-Chat
- Qwen3-8B-Base
- Qwen3Guard-Gen-8B
- Qwen1.5-14B
- Qwen1.5-14B-Chat
- Qwen1.5-MoE-A2.7B
- Qwen1.5-MoE-A2.7B-Chat
- Qwen3-14B-Base
- Qwen2.5-14B
- Qwen2.5-14B-Instruct
- Qwen2.5-14B-Instruct-1M
- Qwen2.5-Coder-14B-Instruct
- Qwen3-30B-A3B
- Qwen3-30B-A3B-Base
- Qwen3-30B-A3B-Instruct-2507
- Qwen3-30B-A3B-Thinking-2507
- Qwen3-Coder-30B-A3B-Instruct
- Qwen1.5-32B
- Qwen1.5-32B-Chat
- QwQ-32B
- QwQ-32B-Preview
- Qwen2.5-32B
- Qwen2.5-Coder-32B-Instruct
- Qwen2-57B-A14B
- Qwen2-57B-A14B-Instruct
- Qwen1.5-72B
- Qwen1.5-72B-Chat
- Qwen2-72B
- Qwen2-72B-Instruct
- Qwen2.5-72B
- Qwen3-235B-A22B
- Qwen3-235B-A22B-Instruct-2507
- Qwen3-235B-A22B-Thinking-2507
- Qwen3-Coder-480B-A35B-Instruct
- SmolLM-135M
- SmolLM-135M-Instruct
- SmolLM2-135M
- SmolLM2-135M-Instruct
- SmolLM2-360M
- SmolLM2-360M-Instruct
- SmolLM-1.7B
- SmolLM-1.7B-Instruct
- SmolLM2-1.7B
- SmolLM2-1.7B-Instruct
- SmolLM3-3B
- SmolLM3-3B-Base
- Yi-1.5-6B
- Yi-6B
- Yi-6B-Chat
- Yi-6B-200K
- Yi-1.5-9B
- Yi-1.5-9B-Chat
- Yi-1.5-9B-Chat-16K
- Yi-9B
- Yi-Coder-9B
- Yi-Coder-9B-Chat
- Yi-1.5-34B
- Yi-1.5-34B-Chat
- Yi-34B
- Yi-34B-Chat
- Yi-34B-200K
- EleutherAI_pythia-6.9b-deduped__sft__tldr
- mistral-7b-anthropic
- mistral-7b-grok
- zephyr-7b-alpha
- zephyr-7b-beta
- KD-Tinker
- zephyr-7b-gemma-sft-v0.1
- zephyr-7b-gemma-v0.1
- zephyr-orpo-141b-A35b-v0.1
- Meta-Llama-3.1-8B-Instruct
- SmolLM-360M-Instruct
- SmolLM-360M
- SmolLM2-1.7B-Instruct-16k
- Yi-9B-200K
- SmolLM3-3B-GSM8K-SFT
- Yi-1.5-34B-32K
- Yi-1.5-9B-32K
- Meta-Llama-3-70B-Instruct
- Yi-1.5-34B-Chat-16K
- Hermes-2-Pro-Llama-3-8B
- Yi-1.5-6B-Chat
- Llama-2-13b-chat-hf
- Yi-Coder-1.5B-Chat
- DeepHermes-3-Llama-3-8B-Preview
- DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos
- DeepHermes-3-Mistral-24B-Preview
- Meta-Llama-3.1-70B-Instruct
- granite-3.1-1b-a400m-base
- Hermes-3-Llama-3.1-70B
- deepseek-coder-7b-base-v1.5
- granite-3.3-2b-base
- Hermes-2-Pro-Mistral-7B
- granite-guardian-3.2-3b-a800m
- Hermes-3-Llama-3.1-405B
- ESFT-token-summary-lite
- granite-guardian-3.2-5b
- Hermes-2-Theta-Llama-3-70B
- ESFT-token-code-lite
- granite-3.0-2b-instruct
- Llama-2-70b-hf
- pythia-410m-subtraction-first-ft
- ESFT-token-translation-lite
- granite-8b-code-base-4k
- Llama-2-70b-chat-hf
- ESFT-token-law-lite
- Olmo-3.1-7B-RL-Zero-Math
- granite-3.0-2b-base
- Meta-Llama-3-70B
- pythia-410m-authors-first-ft
- ESFT-token-math-lite
- granite-3.3-8b-math-prm-v2
- Nous-Hermes-Llama2-13b
- pythia-410m-nli-first-ft
- ESFT-vanilla-lite
- OLMo-2-0425-1B-early-training
- granite-guardian-3.1-2b
- Meta-Llama-3.1-70B
- ESFT-token-intent-lite
- Olmo-3-7B-RL-Zero-IF
- granite-3.0-8b-base
- NousCoder-14B
- ESFT-gate-math-lite
- Olmo-3.1-7B-RL-Zero-Code
- granite-4.1-3b-llmon-preview
- Nous-Hermes-llama-2-7b
- ESFT-gate-code-lite
- OLMo-7B-0724-hf
- granite-8b-code-instruct-128k
- DeepHermes-3-Llama-3-3B-Preview
- ESFT-gate-translation-lite
- OLMo-7B-0724-Instruct-hf
- granite-3.2-8b-instruct-preview
- k2-merged-3.5T-bf16
- ESFT-gate-summary-lite
- OLMo-7B-0424-hf
- granite-guardian-3.0-2b
- Nous-Hermes-2-Mixtral-8x7B-SFT
- FLUX.1 [dev]
- FLUX.1 [schnell]
- Stable Diffusion XL Base 1.0
- Stable Diffusion 3.5 Large
- Wan 2.1 T2V 14B
- Wan 2.1 T2V 1.3B
- CogVideoX 5B
- SDXL Turbo
- Playground v2 512px Base
- Playground v2 1024px Aesthetic
- Playground v2.5 1024px Aesthetic
- RealVisXL V4.0
- RealVisXL V5.0
- DreamShaper XL 1.0
- DreamShaper XL v2 Turbo
- Juggernaut XL v9
- SDXL-Lightning-1step
- SDXL-Lightning-2step
- SDXL-Lightning-4step
- SDXL-Lightning-8step
- Latent Consistency SDXL
- FLUX.1 Krea [dev]
- Stable Diffusion 3.5 Large Turbo
- Qwen-Image
- Qwen-Image-2512
- HunyuanImage 2.1
- HunyuanImage 2.1 Distilled
- Z-Image-Turbo
- CogVideoX 2B
- Mochi 1 Preview
- HunyuanVideo T2V 720p
- HunyuanVideo 1.5 T2V 480p
- HunyuanVideo 1.5 T2V 720p
- HunyuanVideo 1.5 T2V 480p CFG Distilled
- Wan 2.2 TI2V 5B
- LTX-Video 0.9.1 (2B)
- LTX-Video 0.9.5 (2B)
- LTX-Video 0.9.7-dev (13B)
- LTX-Video 0.9.7-distilled (13B)
- LTX-Video 0.9.8-13B-distilled (13B)