Home Price vs performance Value
LLM price vs performance 17 models sit on the price-performance frontier as of October 2026: no other model is both cheaper and better. It runs from gpt-oss-20b ($0.036 blended, 32.5) to GPT-6 Astra ($20, 70.8).
Last verified October 10, 2026
Noometry Index against blended price per million tokens 20 30 40 50 60 70 80 $0.01 $0.1 $1 $10 $100 $1000 gpt-oss-20b: 32.5 at $0.04 gpt-oss-20b Granite 4.0 Micro: 29.0 at $0.04 Gemma 3 4B: 28.1 at $0.05 Qwen3.7 Flash: 39.9 at $0.06 Qwen3.7 Flash Llama 3.1-8B: 23.0 at $0.06 Mistral Small 3: 31.2 at $0.06 Amazon Nova Micro: 30.4 at $0.06 Mercury 2.5: 33.5 at $0.07 gpt-oss-120b: 36.3 at $0.07 Llama 3.2 1B: 20.1 at $0.07 Gemma 3 12B: 32.1 at $0.08 Nemotron 3 Nano 30B A3B: 40.6 at $0.09 Nemotron 3 Nano 30B A3B Nemotron 3.5 Lightning: 40.0 at $0.09 Phi-4: 31.2 at $0.09 Qwen Turbo: 27.1 at $0.09 Gemma 3 27B: 30.8 at $0.10 Ministral 3B: 26.2 at $0.10 Amazon Nova Lite: 31.9 at $0.10 Gemma 4 26B A4B IT: 43.5 at $0.11 Gemma 4 26B A4B IT Granite 4.2 8B: 40.5 at $0.11 Qwen3.5-9B: 33.8 at $0.11 Llama 3.2 3B: 28.9 at $0.12 Phi-4 Mini: 30.9 at $0.13 Mistral Small 3.2: 31.2 at $0.13 GPT-5 Nano: 33.5 at $0.14 Hy3: 44.2 at $0.14 Hy3 GLM-4.7-Flash: 38.8 at $0.15 Ministral 8B: 28.2 at $0.15 Mistral Nemo: 26.4 at $0.15 Devstral Small 2505: 34.3 at $0.15 Llama 4 Scout: 27.7 at $0.15 Step 3.5 Flash: 42.3 at $0.15 Gemma 4 31B IT: 43.5 at $0.15 Llama-3.3-70B-Instruct: 30.6 at $0.16 Nemotron 3 Super: 40.1 at $0.17 Gemini 2.5 Flash-Lite: 37.0 at $0.18 GPT-4.1 nano: 27.9 at $0.18 MiMo-V2-Flash: 41.3 at $0.18 MiMo-V2-Omni: 43.6 at $0.18 MiMo-V2.5: 43.4 at $0.18 MiMo-V2.6-Flash: 48.5 at $0.18 MiMo-V2.6-Flash Qwen3.5-Flash: 42.5 at $0.18 Claude Haiku 5.5: 49.5 at $0.20 Claude Haiku 5.5 GPT-6 Luna: 53.3 at $0.20 GPT-6 Luna Qwen3-30B-A3B: 38.9 at $0.21 GLM-5.3-Flash: 51.8 at $0.24 Mistral 7B: 23.0 at $0.25 Command R: 31.4 at $0.26 DeepSeek V4 Flash: 53.6 at $0.26 DeepSeek V4 Flash DeepSeek V4.1 Flash: 52.8 at $0.26 GPT-4o mini: 25.5 at $0.26 Mistral Small: 33.4 at $0.26 Grok 4.1 Fast: 41.4 at $0.28 DeepSeek-V3.2-Exp: 44.3 at $0.29 Qwen3 Coder Next: 34.3 at $0.29 Llama 4 Maverick: 30.9 at $0.30 Qwen2.5 7B Instruct: 29.0 at $0.31 Qwen3 8B: 33.7 at $0.31 Mercury 2: 39.1 at $0.38 Mistral Large 3: 39.1 at $0.38 Trinity Large Thinking: 38.6 at $0.39 Llama 3.1-70B: 29.6 at $0.40 Nvidia Llama 3.3 Nemotron Super 49b v1.5: 40.3 at $0.40 Mistral Small 3.1: 31.7 at $0.40 DeepSeek-V3: 39.5 at $0.41 Step 3.7 Flash: 37.3 at $0.42 Qwen3.6 Flash: 38.8 at $0.42 DeepSeek-V3.1: 42.8 at $0.42 GLM-4.5-Air: 38.9 at $0.43 Codestral: 30.6 at $0.45 GLM-4.6V: 41.3 at $0.45 GPT-5.6 Luna: 54.6 at $0.45 GPT-5.6 Luna DeepSeek-V3.1-Terminus: 43.1 at $0.45 GPT-5.4 nano: 41.9 at $0.46 MiniMax-M2: 37.4 at $0.52 MiniMax-M2.1: 38.9 at $0.52 MiniMax-M2.5: 38.3 at $0.52 MiniMax-M2.7: 37.7 at $0.52 MiniMax-M3: 43.8 at $0.52 Solar Pro4: 42.1 at $0.52 MiMo-V2-Pro: 43.0 at $0.54 MiMo-V2.5-Pro: 45.2 at $0.54 MiMo-V2.6-Pro: 50.3 at $0.54 Qwen3.6 35B-A3B: 37.6 at $0.56 Gemini 3.1 Flash Lite: 40.8 at $0.56 Qwen Plus: 37.1 at $0.60 Qwen3 14B: 35.5 at $0.61 Inkling-Small: 46.5 at $0.64 Gemma 2 27B: 29.4 at $0.65 GPT-5 Mini: 41.8 at $0.69 Qwen3.5 35B-A3B: 42.0 at $0.69 Mixtral 8x7B: 27.1 at $0.70 GPT-4.1 mini: 33.6 at $0.70 Qwen3.7 Plus: 45.3 at $0.70 Qwen2.5-Coder-32B: 33.4 at $0.74 GPT-3.5-turbo: 23.2 at $0.75 Magistral Small: 30.2 at $0.75 Mistral Medium 3.1: 31.9 at $0.80 Qwen3.5 27B: 41.9 at $0.82 Gemini 2.5 Flash: 39.3 at $0.85 Gemini 3.5 Flash Lite: 41.5 at $0.85 Nova 2 Lite: 39.7 at $0.85 DeepSeek-V3.2-Speciale: 39.7 at $0.85 Qwen3-Next 80B-A3B Instruct: 43.0 at $0.88 GLM-4.5V: 39.8 at $0.90 Kimi K2.5: 48.1 at $0.90 Qwen3.5 Plus: 42.9 at $0.90 DeepSeek-R1: 42.3 at $0.91 Nemotron 3 Ultra: 42.5 at $0.93 MiniMax M1: 40.3 at $0.96 DeepSeek V4 Pro: 54.3 at $0.99 GLM-4.5: 42.0 at $1.00 GLM-4.6: 41.4 at $1.00 GLM-4.7: 42.0 at $1.00 Sonar: 38.5 at $1.00 Kimi K2 (Jul 2025): 41.2 at $1.00 Mistral Large 4: 43.1 at $1.03 Qwen3.5 122B-A10B: 42.1 at $1.10 Qwen3.8 27B: 46.0 at $1.11 Gemini 3 Flash Preview: 52.3 at $1.13 Qwen3.6 Plus: 47.5 at $1.13 Seed 2.0 Pro: 43.2 at $1.13 Hy4 preview: 45.3 at $1.13 Qwen2.5 32B Instruct: 30.1 at $1.22 Qwen3 235B-A22B: 43.5 at $1.22 Qwen3 32B: 39.2 at $1.22 Qwen3-VL 235B-A22B: 43.2 at $1.22 Grok Build 0.1: 36.4 at $1.25 Qwen3.5 397B-A17B: 46.0 at $1.35 Qwen3.6 27B: 42.2 at $1.35 Amazon Nova Pro: 31.0 at $1.40 Step 5 Preview: 47.9 at $1.43 Gemini 3.6 Flash: 54.1 at $1.50 Gemini 3.7 Flash: 59.8 at $1.50 Gemini 3.7 Flash Gemini 3.8 Flash: 61.8 at $1.50 Gemini 3.8 Flash GLM-5: 46.1 at $1.55 Grok 4.20 (Non-Reasoning): 48.6 at $1.56 Grok 4.20 Multi-Agent: 46.2 at $1.56 Grok 4.3: 43.8 at $1.56 GPT-5.4 mini: 45.0 at $1.69 Kimi K2.6: 47.7 at $1.71 Kimi K2.7 Code: 43.3 at $1.71 GLM-5V-Turbo: 43.8 at $1.90 o3-mini: 36.7 at $1.93 o4-mini: 41.6 at $1.93 Claude Haiku 4.5: 39.5 at $2.00 Muse Spark 1.1: 49.9 at $2.00 Muse Spark 1.2: 50.3 at $2.00 Muse Spark 1.3: 54.8 at $2.00 GLM-5.1: 47.8 at $2.15 GLM-5.2: 51.1 at $2.15 GLM-5.3: 54.8 at $2.15 Qwen3 Max: 43.7 at $2.40 Qwen2.5 72B Instruct: 31.9 at $2.45 Inkling: 44.1 at $2.57 Magistral Medium: 35.2 at $2.75 Qwen Max: 34.7 at $2.80 Qwen3.6 Max Preview: 51.5 at $2.92 Grok 4.5: 55.0 at $3.00 Grok 4.6: 56.9 at $3.00 Grok 4.7: 53.1 at $3.00 Mistral Large: 31.9 at $3.00 Mistral Medium: 36.3 at $3.00 Mistral Medium 3.5: 40.2 at $3.00 Mixtral 8x22B: 27.1 at $3.00 Pixtral Large: 32.2 at $3.00 Qwen3-Coder 480B-A35B Instruct: 38.1 at $3.00 Qwen3.8 Max: 56.8 at $3.00 Gemini 3.5 Flash: 54.2 at $3.38 Gemini 2.5 Pro: 45.0 at $3.44 GPT-5: 50.9 at $3.44 GPT-5-Codex: 37.9 at $3.44 GPT-5.1: 49.0 at $3.44 GPT-5.1-Codex: 38.6 at $3.44 GPT-4.1: 35.9 at $3.50 o3: 47.5 at $3.50 Qwen3.7 Max: 51.5 at $3.75 Claude Sonnet 5: 54.6 at $4.00 Claude Sonnet 5.5: 61.9 at $4.00 Claude Sonnet 5.5 GPT-6 Sol: 61.8 at $4.00 GPT-6.1 Sol: 65.6 at $4.00 GPT-6.1 Sol Qwen2.5-VL 72B Instruct: 29.9 at $4.20 Command A: 36.5 at $4.38 Command R+: 32.4 at $4.38 GPT-4o: 28.6 at $4.38 Gemini 3.1 Pro Preview: 56.7 at $4.50 GPT-5.6 Terra: 59.2 at $4.50 GPT-5.2: 54.1 at $4.81 GPT-5.2 Codex: 42.6 at $4.81 GPT-5.3 Chat: 42.8 at $4.81 GPT-5.3 Codex: 45.8 at $4.81 Nova Premier 1.0: 38.3 at $5.00 GPT-5.4: 59.4 at $5.63 Claude Sonnet 4: 40.8 at $6.00 Claude Sonnet 4.5: 44.1 at $6.00 Claude Sonnet 4.6: 50.3 at $6.00 Kimi K3: 59.5 at $6.00 Claude Opus 5.5: 68.6 at $8.00 Claude Opus 5.5 GPT-5.6 Sol: 65.0 at $8.00 Claude Opus 4.5: 50.5 at $10.00 Claude Opus 4.6: 58.2 at $10.00 Claude Opus 4.7: 58.3 at $10.00 Claude Opus 4.8: 60.7 at $10.00 Claude Opus 5: 67.8 at $10.00 GPT-5.5: 63.4 at $11.25 GPT-4 Turbo: 30.5 at $15.00 Claude Fable 5: 66.8 at $20.00 Claude Fable 5.1: 69.0 at $20.00 Claude Fable 5.1 GPT-6 Astra: 70.8 at $20.00 GPT-6 Astra o1: 40.9 at $26.25 Claude Opus 4: 43.1 at $30.00 Claude Opus 4.1: 41.0 at $30.00 o3-pro: 42.9 at $35.00 GPT-4: 29.1 at $37.50 GPT-5 Pro: 46.4 at $41.25 GPT-5.2 Pro: 52.3 at $57.75 GPT-5.4 Pro: 58.9 at $67.50 GPT-5.5 Pro: 64.3 at $67.50 o1-pro: 31.5 at $262.50 Blended $ per 1M tokens (log scale) Noometry Index Frequently asked questions Which AI model is the best value? 17 models sit on the price-performance frontier as of October 2026: no other model is both cheaper and better. It runs from gpt-oss-20b ($0.036 blended, 32.5) to GPT-6 Astra ($20, 70.8).
What is the price-performance frontier? The set of models that no other model beats on both price and score. If a model is not on it, something cheaper scores at least as well.
Cite this page Noometry. (2026). LLM price vs performance. Retrieved October 10, 2026, from https://noometry.com/llm-value
Quote Noometry with a link back to this page. It is also available in Markdown .