Intelligence Per Dollar
Benchmark points per blended dollar across every credibly ranked model. Scores are min-max normalized across 233 models with 4+ independent benchmarks; cost assumes a typical 3:1 input:output token mix. Leader = 100.
This is not a quality ranking. #1 here means the most benchmark points per dollar, so a mid-scoring budget model will outrank a frontier model that costs 100x more. Check the Score and % of top score columns for raw capability, and use the task rankings when output quality is what compounds in your workflow.
Value leaderboard
| # | Model | Score | % of top score | In / Out per 1M | Blended $/1M | Value |
|---|---|---|---|---|---|---|
| 1 | Ling-3.0-flash BEST VALUEinclusionai/ling-3.0-flash | 64.4 | 64% | $0.02 / $0.06 | $0.03 | 100.0 |
| 2 | inclusionAI: Ling-2.6-flashinclusionai/ling-2.6-flash | 21.6 | 22% | $0.01 / $0.03 | $0.01 | 70.4 |
| 3 | Upstage: Solar Pro 4upstage/solar-pro4 | 70.2 | 70% | $0.03 / $0.12 | $0.05 | 65.4 |
| 4 | OpenAI: gpt-oss-120bopenai/gpt-oss-120b | 43.1 | 43% | $0.03 / $0.17 | $0.07 | 32.4 |
| 5 | Mistral: Mistral Small 3mistralai/mistral-small-24b-instruct-2501 | 28.8 | 29% | $0.05 / $0.08 | $0.06 | 24.5 |
| 6 | OpenAI: gpt-oss-20bopenai/gpt-oss-20b | 25.7 | 26% | $0.03 / $0.13 | $0.06 | 22.9 |
| 7 | OpenAI: GPT-5 Nano (batch)openai/gpt-5-nano:batch | 32.1 | 32% | $0.03 / $0.20 | $0.07 | 22.8 |
| 8 | Mistral: Mistral Small 3.2 24Bmistralai/mistral-small-3.2-24b-instruct | 59.6 | 60% | $0.09 / $0.25 | $0.13 | 21.9 |
| 9 | OpenAI: GPT-5.6 Lunaopenai/gpt-5.6-luna | 88.2 | 88% | $0.10 / $0.60 | $0.23 | 19.2 |
| 10 | OpenAI: GPT-5.6 Luna (batch)openai/gpt-5.6-luna:batch | 88.2 | 88% | $0.10 / $0.60 | $0.23 | 19.2 |
| 11 | Google: Gemma 4 31Bgoogle/gemma-4-31b-it | 56.7 | 57% | $0.10 / $0.34 | $0.16 | 17.3 |
| 12 | Z.ai: GLM 4.7 Flashz-ai/glm-4.7-flash | 46.7 | 47% | $0.06 / $0.40 | $0.15 | 15.8 |
| 13 | OpenAI: GPT-5.4 Nano (batch)openai/gpt-5.4-nano:batch | 68.6 | 69% | $0.10 / $0.62 | $0.23 | 14.5 |
| 14 | MiniMax: MiniMax M3 (batch)minimax/minimax-m3:batch | 77.0 | 77% | $0.15 / $0.60 | $0.26 | 14.3 |
| 15 | StepFun: Step 3.5 Flashstepfun/step-3.5-flash | 42.9 | 43% | $0.10 / $0.30 | $0.15 | 14.0 |
| 16 | inclusionAI: Ring-2.6-1Tinclusionai/ring-2.6-1t | 58.8 | 59% | $0.07 / $0.62 | $0.21 | 13.5 |
| 17 | Qwen: Qwen3.5-9Bqwen/qwen3.5-9b | 30.8 | 31% | $0.10 / $0.15 | $0.11 | 13.4 |
| 18 | Google: Gemma 4 26B A4B google/gemma-4-26b-a4b-it | 51.7 | 52% | $0.12 / $0.40 | $0.19 | 13.3 |
| 19 | Qwen: Qwen3 32Bqwen/qwen3-32b | 35.0 | 35% | $0.08 / $0.28 | $0.13 | 13.2 |
| 20 | NVIDIA: Nemotron 3.5 Lightningnvidia/nemotron-3.5-lightning | 34.6 | 35% | $0.10 / $0.25 | $0.14 | 12.3 |
| 21 | NVIDIA: Nemotron 3 Supernvidia/nemotron-3-super-120b-a12b | 39.0 | 39% | $0.09 / $0.40 | $0.16 | 11.6 |
| 22 | OpenAI: GPT-5 Nanoopenai/gpt-5-nano | 32.1 | 32% | $0.05 / $0.40 | $0.14 | 11.4 |
| 23 | Google: Gemma 3 4Bgoogle/gemma-3-4b-it | 14.5 | 14% | $0.05 / $0.10 | $0.06 | 11.3 |
| 24 | inclusionAI: Ling-2.6-1Tinclusionai/ling-2.6-1t | 42.9 | 43% | $0.07 / $0.62 | $0.21 | 9.9 |
| 25 | DeepSeek: DeepSeek V3deepseek/deepseek-chat | 86.3 | 86% | $0.26 / $1.03 | $0.45 | 9.4 |
| 26 | Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro | 81.1 | 81% | $0.25 / $1.00 | $0.44 | 9.1 |
| 27 | OpenAI: GPT-4.1 Nano (batch)openai/gpt-4.1-nano:batch | 15.4 | 15% | $0.05 / $0.20 | $0.09 | 8.6 |
| 28 | Qwen: Qwen3 Coder 30B A3B Instructqwen/qwen3-coder-30b-a3b-instruct | 21.2 | 21% | $0.07 / $0.28 | $0.12 | 8.5 |
| 29 | IBM: Granite 4.1 8Bibm-granite/granite-4.1-8b | 10.3 | 10% | $0.05 / $0.10 | $0.06 | 8.1 |
| 30 | OpenAI: GPT-5.4 Nanoopenai/gpt-5.4-nano | 68.6 | 69% | $0.20 / $1.25 | $0.46 | 7.3 |
Budget champions : 80+ score, cheapest first
| # | Model | Score | % of top score | In / Out per 1M | Blended $/1M | Value |
|---|---|---|---|---|---|---|
| 1 | OpenAI: GPT-5.6 Lunaopenai/gpt-5.6-luna | 88.2 | 88% | $0.10 / $0.60 | $0.23 | 19.2 |
| 2 | OpenAI: GPT-5.6 Luna (batch)openai/gpt-5.6-luna:batch | 88.2 | 88% | $0.10 / $0.60 | $0.23 | 19.2 |
| 3 | Nex AGI: Nex-N2-Pronex-agi/nex-n2-pro | 81.1 | 81% | $0.25 / $1.00 | $0.44 | 9.1 |
| 4 | DeepSeek: DeepSeek V3deepseek/deepseek-chat | 86.3 | 86% | $0.26 / $1.03 | $0.45 | 9.4 |
| 5 | OpenAI: o4 Mini High (batch)openai/o4-mini-high:batch | 81.0 | 81% | $0.55 / $2.20 | $0.96 | 4.1 |
| 6 | Z.ai: GLM 5.2 (batch)z-ai/glm-5.2:batch | 89.3 | 89% | $0.70 / $2.20 | $1.07 | 4.1 |
| 7 | Z.ai: GLM 5.2z-ai/glm-5.2 | 89.3 | 89% | $0.50 / $3.15 | $1.16 | 3.8 |
| 8 | DeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-pro | 90.5 | 90% | $1.17 / $2.34 | $1.46 | 3.0 |
| 9 | Google: Gemini 3.6 Flash (batch)google/gemini-3.6-flash:batch | 84.4 | 84% | $0.75 / $3.75 | $1.50 | 2.8 |
| 10 | Google: Gemini 3.5 Flash (batch)google/gemini-3.5-flash:batch | 84.3 | 84% | $0.75 / $4.50 | $1.69 | 2.4 |
Free models with credible scores
Per-dollar math breaks at $0. These are simply the strongest free options:
- NVIDIA: Nemotron 3 Ultra (free) : score 63.0
- Google: Gemma 4 31B (free) : score 56.7
- Google: Gemma 4 26B A4B (free) : score 51.7
- NVIDIA: Nemotron 3 Super (free) : score 39.0
- NVIDIA: Nemotron 3.5 Lightning (free) : score 34.6
- OpenAI: gpt-oss-20b (free) : score 25.7
- NVIDIA: Nemotron 3 Nano 30B A3B (free) : score 10.4
Assumptions
Value = blended benchmark score divided by blended price per million tokens, indexed to the leader. A 3:1 input:output ratio fits most chat and RAG workloads; estimate your exact mix with the cost calculator. Scoring details in the methodology. Models with fewer than 4 independent benchmarks are excluded rather than guessed.