# Best Value LLM for Knowledge in 2026 — Cost-Adjusted Rankings

> Top AI models ranked by knowledge benchmark performance per dollar. Cost-adjusted rankings using GPQA, MMLU-Pro, HLE, and more.

Knowledge benchmarks test factual accuracy, expert-level science, and professional comprehension. This ranking divides each model's weighted knowledge score (GPQA, SuperGPQA, MMLU-Pro, HLE, FrontierScience, SimpleQA) by output token price. For RAG pipelines, research assistants, and Q&A systems where accurate knowledge retrieval matters but API costs add up, the value leaders here offer the best accuracy per dollar.

Canonical page: https://benchlm.ai/best/best-value-knowledge

Last updated: October 8, 2026

## Rankings

| Rank | Model | Creator | Type | Context | Score/$ | Score | Output $/1M |
|------|-------|---------|------|---------|---------|-------|-------------|
| 1 | [Mercury 2.5](/models/mercury-2-5) | Inception | Proprietary | 260K | 223 | 33.5 | $0.15 |
| 2 | [MiMo-V2.6-Flash](/models/mimo-v2-6-flash) | Xiaomi | Open Weight | 1M | 216.61 | 60.7 | $0.28 |
| 3 | [GPT-6 Luna](/models/gpt-6-luna) | OpenAI | Proprietary | 1.05M | 128.52 | 64.3 | $0.5 |
| 4 | [Qwen3.5 Flash](/models/qwen3-5-flash) | Alibaba | Proprietary | 1M | 109.98 | 44 | $0.4 |
| 5 | [DeepSeek V3.2](/models/deepseek-v3-2) | DeepSeek | Open Weight | 128K | 99.14 | 41.6 | $0.42 |
| 6 | [MiMo-V2.6-Pro](/models/mimo-v2-6-pro) | Xiaomi | Open Weight | 1M | 78.24 | 68.1 | $0.87 |
| 7 | [GPT-4.1 nano](/models/gpt-4-1-nano) | OpenAI | Proprietary | 1M | 62.13 | 24.9 | $0.4 |
| 8 | [Grok 4.1 Fast](/models/grok-4-1-fast) | xAI | Proprietary | 1M | 61.04 | 30.5 | $0.5 |
| 9 | [Mistral Small 4](/models/mistral-small-4) | Mistral | Open Weight | 256K | 55.12 | 33.1 | $0.6 |
| 10 | [GPT-5.6 Luna](/models/gpt-5-6-luna) | OpenAI | Proprietary | 1.05M | 53.78 | 64.5 | $1.2 |
| 11 | [DeepSeek V4.1 Flash](/models/deepseek-v4-1-flash) | DeepSeek | Open Weight | 1M | 53.65 | 64.4 | $1.2 |
| 12 | [Solar Pro 3](/models/solar-pro-3) | Upstage | Proprietary | 128K | 49.08 | 29.5 | $0.6 |
| 13 | [MiniMax M3](/models/minimax-m3) | MiniMax | Open Weight | 1M | 40.31 | 48.4 | $1.2 |
| 14 | [Step 3.7 Flash](/models/step-3-7-flash) | StepFun | Open Weight | 256K | 38.18 | 43.9 | $1.15 |
| 15 | [Celeris-1](/models/celeris-1) | Celeris | Proprietary | 128K | 38.03 | 26.6 | $0.7 |
| 16 | [Trinity-Large-Thinking](/models/trinity-large-thinking) | Arcee AI | Open Weight | 512K | 37.71 | 33.9 | $0.9 |
| 17 | [Inkling-Small](/models/inkling-small) | Thinking Machines Lab | Open Weight | 1M | 37.31 | 53.7 | $1.44 |
| 18 | [MiniMax M2.7](/models/minimax-m2-7) | MiniMax | Open Weight | 200K | 36.37 | 43.6 | $1.2 |
| 19 | [GPT-4o mini](/models/gpt-4o-mini) | OpenAI | Proprietary | 128K | 34.73 | 20.8 | $0.6 |
| 20 | [Gemini 3.1 Flash-Lite](/models/gemini-3-1-flash-lite) | Google | Proprietary | 1M | 31.63 | 47.5 | $1.5 |
| 21 | [GPT-5.4 nano](/models/gpt-5-4-nano) | OpenAI | Proprietary | 400K | 30.38 | 38 | $1.25 |
| 22 | [GLM-4.5-Air](/models/glm-4-5-air) | Z.AI | Proprietary | 128K | 28.51 | 31.4 | $1.1 |
| 23 | [DeepSeek V3](/models/deepseek-v3) | DeepSeek | Open Weight | 128K | 27.33 | 30.1 | $1.1 |
| 24 | [Mistral Large 4](/models/mistral-large-4) | Mistral | Pending | 1M | 26.33 | 55 | $2.09 |
| 25 | [Quasar 438B](/models/quasar-438b) | Multiverse Computing | Proprietary | 1M | 25.45 | 45.8 | $1.8 |
| 26 | [Step 5 Preview](/models/step-5-preview) | StepFun | Pending | 1M | 24.81 | 67 | $2.7 |
| 27 | [Grok Code Fast 1](/models/grok-code-fast-1) | xAI | Proprietary | 256K | 23.29 | 34.9 | $1.5 |
| 28 | [Grok 4.3](/models/grok-4-3) | xAI | Proprietary | 1M | 21.6 | 54 | $2.5 |
| 29 | [Mistral Large 3](/models/mistral-large-3) | Mistral | Proprietary | 128K | 20.57 | 30.9 | $1.5 |
| 30 | [Grok 4.20](/models/grok-4-20-beta) | xAI | Proprietary | 2M | 20.18 | 50.5 | $2.5 |
| 31 | [GPT-4.1 mini](/models/gpt-4-1-mini) | OpenAI | Proprietary | 1M | 19.46 | 31.1 | $1.6 |
| 32 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | Google | Proprietary | 1M | 19.31 | 72.4 | $3.75 |
| 33 | [Gemini 3.5 Flash-Lite](/models/gemini-3-5-flash-lite) | Google | Proprietary | 1M | 19.31 | 48.3 | $2.5 |
| 34 | [Gemini 3.7 Flash](/models/gemini-3-7-flash) | Google | Proprietary | 1M | 18.72 | 70.2 | $3.75 |
| 35 | [Gemini 3 Flash](/models/gemini-3-flash) | Google | Proprietary | 1M | 17.88 | 53.6 | $3 |
| 36 | [DeepSeek-R1](/models/deepseek-r1) | DeepSeek | Open Weight | 128K | 17.73 | 38.8 | $2.19 |
| 37 | [Gemini 3.6 Flash](/models/gemini-3-6-flash) | Google | Proprietary | 1M | 16.87 | 63.3 | $3.75 |
| 38 | [Kimi K2.5](/models/kimi-k2-5) | Moonshot AI | Open Weight | 256K | 16.6 | 49.8 | $3 |
| 39 | [Claude 3 Haiku](/models/claude-3-haiku) | Anthropic | Proprietary | 200K | 15.84 | 19.8 | $1.25 |
| 40 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | Open Weight | 1M | 15.83 | 62.7 | $3.96 |
| 41 | [GLM-5](/models/glm-5) | Z.AI | Open Weight | 200K | 15.15 | 48.5 | $3.2 |
| 42 | [Muse Spark 1.2](/models/muse-spark-1-2) | Meta | Proprietary | 1M | 15.08 | 64.1 | $4.25 |
| 43 | [Mistral Medium 3](/models/mistral-medium-3) | Mistral | Proprietary | 128K | 14.91 | 29.8 | $2 |
| 44 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | Open Weight | 256K | 14.56 | 58.2 | $4 |
| 45 | [Gemini 2.5 Flash](/models/gemini-2-5-flash) | Google | Proprietary | 1M | 14.38 | 35.9 | $2.5 |
| 46 | [Kimi K2](/models/kimi-k2) | Moonshot AI | Proprietary | 128K | 14.36 | 35.9 | $2.5 |
| 47 | [Qwen3.5 397B](/models/qwen3-5-397b) | Alibaba | Open Weight | 128K | 13.78 | 49.6 | $3.6 |
| 48 | [Kimi K2.7 Code](/models/kimi-k2-7-code) | Moonshot AI | Open Weight | 256K | 13.42 | 53.7 | $4 |
| 49 | [GLM-5.2](/models/glm-5-2) | Z.AI | Open Weight | 1M | 12.95 | 57 | $4.4 |
| 50 | [GLM-5.1](/models/glm-5-1) | Z.AI | Open Weight | 203K | 11.97 | 52.7 | $4.4 |
| 51 | [GLM-5-Turbo](/models/glm-5-turbo) | Z.AI | Proprietary | 200K | 11.91 | 47.6 | $4 |
| 52 | [Inkling](/models/inkling) | Thinking Machines Lab | Open Weight | 1M | 11.56 | 54.1 | $4.68 |
| 53 | [Grok 4.6](/models/grok-4-6) | xAI | Proprietary | 500K | 11.38 | 68.3 | $6 |
| 54 | [GLM-5V-Turbo](/models/glm-5v-turbo) | Z.AI | Proprietary | 200K | 11.24 | 44.9 | $4 |
| 55 | [Grok 4.5](/models/grok-4-5) | xAI | Proprietary | 500K | 11.17 | 67 | $6 |
| 56 | [GPT-5.4 mini](/models/gpt-5-4-mini) | OpenAI | Proprietary | 400K | 10.78 | 48.5 | $4.5 |
| 57 | [GPT-6.1 Sol](/models/gpt-6-1-sol) | OpenAI | Proprietary | 1.05M | 8.27 | 82.7 | $10 |
| 58 | [Claude Sonnet 5.5](/models/claude-sonnet-5-5) | Anthropic | Proprietary | 1M | 8.15 | 81.5 | $10 |
| 59 | [o3-mini](/models/o3-mini) | OpenAI | Proprietary | 200K | 8.05 | 35.4 | $4.4 |
| 60 | [Gemini 4 Argon](/models/gemini-4-argon) | Google | Proprietary | null | 7.82 | 78.2 | $10 |
| 61 | [GPT-6 Sol](/models/gpt-6-sol) | OpenAI | Proprietary | 1.05M | 7.67 | 76.7 | $10 |
| 62 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | Google | Proprietary | 1M | 7.06 | 63.5 | $9 |
| 63 | [Claude Haiku 4.5](/models/claude-haiku-4-5) | Anthropic | Proprietary | 200K | 6.93 | 34.7 | $5 |
| 64 | [Claude Sonnet 5](/models/claude-sonnet-5) | Anthropic | Proprietary | 1M | 6.4 | 64 | $10 |
| 65 | [o3](/models/o3) | OpenAI | Proprietary | 200K | 5.89 | 47.1 | $8 |
| 66 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | OpenAI | Proprietary | 1.05M | 5.86 | 70.3 | $12 |
| 67 | [Gemini 3.1 Pro](/models/gemini-3-1-pro) | Google | Proprietary | 1M | 5.48 | 65.8 | $12 |
| 68 | [Gemini 1.5 Pro](/models/gemini-1-5-pro) | Google | Proprietary | 2M | 5.29 | 26.5 | $5 |
| 69 | [GPT-5.1](/models/gpt-5-1) | OpenAI | Proprietary | 200K | 5.23 | 52.3 | $10 |
| 70 | [GPT-5.1-Codex](/models/gpt-5-1-codex) | OpenAI | Proprietary | 400K | 5.05 | 50.5 | $10 |
| 71 | [Gemini 3 Pro](/models/gemini-3-pro) | Google | Proprietary | 2M | 4.91 | 58.9 | $12 |
| 72 | [GPT-5.3 Codex](/models/gpt-5-3-codex) | OpenAI | Proprietary | 400K | 4.59 | 64.3 | $14 |
| 73 | [Kimi K3](/models/kimi-k3) | Moonshot AI | Pending | 1.05M | 4.49 | 67.4 | $15 |
| 74 | [GPT-4.1](/models/gpt-4-1) | OpenAI | Proprietary | 1M | 4.49 | 35.9 | $8 |
| 75 | [Claude Opus 5.5](/models/claude-opus-5-5) | Anthropic | Proprietary | 1M | 4.42 | 88.4 | $20 |
| 76 | [Gemini 2.5 Pro](/models/gemini-2-5-pro) | Google | Proprietary | 1M | 4.4 | 44 | $10 |
| 77 | [GPT-5.4](/models/gpt-5-4) | OpenAI | Proprietary | 1.05M | 4.39 | 65.8 | $15 |
| 78 | [GPT-5.2](/models/gpt-5-2) | OpenAI | Proprietary | 400K | 4.14 | 58 | $14 |
| 79 | [Mistral Medium 3.5 128B](/models/mistral-medium-3-5-128b) | Mistral | Open Weight | 256K | 4.11 | 30.8 | $7.5 |
| 80 | [GPT-5.2-Codex](/models/gpt-5-2-codex) | OpenAI | Proprietary | 400K | 3.96 | 55.5 | $14 |
| 81 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | OpenAI | Proprietary | 1.05M | 3.89 | 77.7 | $20 |
| 82 | [Claude Sonnet 4.6](/models/claude-sonnet-4-6) | Anthropic | Proprietary | 200K | 3.57 | 53.6 | $15 |
| 83 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | Proprietary | null | 3.21 | 80.3 | $25 |
| 84 | [GPT-4o](/models/gpt-4o) | OpenAI | Proprietary | 128K | 2.81 | 28.1 | $10 |
| 85 | [Command A+](/models/command-a-plus) | Cohere | Open Weight | 128K | 2.81 | 28.1 | $10 |
| 86 | [Claude Opus 4.8](/models/claude-opus-4-8) | Anthropic | Proprietary | 1M | 2.77 | 69.3 | $25 |
| 87 | [Claude Opus 4.7 (Adaptive)](/models/claude-opus-4-7-adaptive) | Anthropic | Proprietary | 1M | 2.56 | 64 | $25 |
| 88 | [Claude Opus 4.7](/models/claude-opus-4-7) | Anthropic | Proprietary | 1M | 2.55 | 63.7 | $25 |
| 89 | [Claude 4 Sonnet](/models/claude-4-sonnet) | Anthropic | Proprietary | 200K | 2.44 | 36.7 | $15 |
| 90 | [Claude Opus 4.6](/models/claude-opus-4-6) | Anthropic | Proprietary | 1M | 2.33 | 58.3 | $25 |
| 91 | [GPT-5.5](/models/gpt-5-5) | OpenAI | Proprietary | 1M | 2.31 | 69.3 | $30 |
| 92 | [Claude Opus 4.5](/models/claude-opus-4-5) | Anthropic | Proprietary | 200K | 1.97 | 49.4 | $25 |
| 93 | [GPT-6 Astra](/models/gpt-6-astra) | OpenAI | Proprietary | 1.05M | 1.7 | 85 | $50 |
| 94 | [Claude Fable 5.1](/models/claude-fable-5-1) | Anthropic | Proprietary | 1M | 1.7 | 85 | $50 |
| 95 | [Claude Fable 5](/models/claude-fable) | Anthropic | Proprietary | 1M+ | 1.62 | 80.8 | $50 |
| 96 | [o1](/models/o1) | OpenAI | Proprietary | 200K | 0.68 | 40.9 | $60 |
| 97 | [Claude 3 Opus](/models/claude-3-opus) | Anthropic | Proprietary | 200K | 0.32 | 23.7 | $75 |
| — | [Mistral Large 4](/models/mistral-large-4) | Mistral | Pending | 1M | — | — | — |

## Key Takeaways

- Top model: [Mercury 2.5](/models/mercury-2-5) with a Score/$ of 223 (score 33.5, $0.15/1M output tokens)
- Best open-weight option: [MiMo-V2.6-Flash](/models/mimo-v2-6-flash) at #2
- Models included: 98 (97 ranked, 1 listed without a rank)

## Compare the Leaders

- [Mercury 2.5 vs MiMo-V2.6-Flash](/compare/mercury-2-5-vs-mimo-v2-6-flash)