LLM leaderboard
LLM Leaderboard — Model Rankings by Price & Performance
Independent ranking of over 1,000 frontier and open-weight language models, scored by Value Score — quality benchmark divided by blended $/1M token cost, normalized across every tracked model. Token usage, list pricing, context window, and 30-day price moves are refreshed daily so you can see which models deliver the most performance per dollar this week.
Last updated: 2026-11-02 · 86 models · sorted by Value Score (highest first)
Value Score formula: Value Score = quality benchmark ÷ blended token cost, normalized across all models. Higher = more performance per dollar.
| # | Model | Provider | Category | Input $/M | Output $/M | Context | Benchmark | Value Score | Tokens | DoD | Price Δ 30d |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 01 | Ministral 8B ministral-8b |
Mistral | Budget | $0.100 | $0.100 | 128K | 58 | 100 High Value | 152B | +12% | 0% |
| 02 | MiMo V2.5 Lite mimo-v2-5-lite |
Xiaomi | Budget | $0.050 | $0.200 | 64K | 58 | 71 Fair Value | 215B | +16% | 0% |
| 03 | DeepSeek V4 Flash deepseek-v4-flash |
DeepSeek | Budget | $0.070 | $0.280 | 128K | 78 | 69 Fair Value | 5.47T | +95% | -30% |
| 04 | MiMo V2.6 Flash mimo-v2-6-flash |
Xiaomi | Budget | $0.140 | $0.280 | 1.1M | 86 | 66 Fair Value | 113B | +1006% | 0% |
| 05 | DeepSeek V3.2 deepseek-v3-2 |
DeepSeek | Budget | $0.140 | $0.280 | 128K | 75 | 58 Fair Value | 1.06T | -15% | -50% |
| 06 | abab6.5s-chat minimax-abab-6-5s-chat |
MiniMax | Budget | $0.100 | $0.300 | 128K | 60 | 47 Fair Value | 126B | +17% | 0% |
| 07 | Gemini 3 Flash gemini-3-flash |
Budget | $0.100 | $0.400 | 1M | 74 | 46 Fair Value | 465B | +22% | -33% | |
| 08 | GPT-6 Luna gpt-6-luna |
OpenAI | Budget | $0.100 | $0.500 | 1.1M | 86 | 44 Fair Value | 121B | +1011% | 0% |
| 09 | Hy3 Lite hy3-lite |
Tencent | Budget | $0.100 | $0.400 | 128K | 64 | 39 Low Value | 162B | +29% | 0% |
| 10 | Kimi K2 Mini kimi-k2-mini |
Moonshot | Budget | $0.100 | $0.400 | 128K | 60 | 37 Low Value | 171B | +11% | 0% |
| 11 | Gemini 2.5 Flash gemini-2-5-flash |
Budget | $0.150 | $0.600 | 1M | 70 | 29 Low Value | 577B | -8% | 0% | |
| 12 | MiniMax M2 minimax-m2 |
MiniMax | Coding | $0.200 | $0.800 | 200K | 74 | 23 Low Value | 247B | +42% | 0% |
| 13 | MiMo V2.5 mimo-v2-5 |
Xiaomi | Open | $0.200 | $0.800 | 128K | 71 | 22 Low Value | 2.72T | +52% | 0% |
| 14 | MiMo V2.6 Pro mimo-v2-6-pro |
Xiaomi | Flagship | $0.435 | $0.870 | 1.1M | 90 | 22 Low Value | 118B | +1023% | 0% |
| 15 | MiniMax M1 minimax-m1 |
MiniMax | Reasoning | $0.200 | $0.800 | 128K | 68 | 21 Low Value | 201B | 0% | 0% |
| 16 | Llama 4 Maverick llama-4-maverick |
Meta | Open | $0.270 | $0.850 | 1M | 73 | 20 Low Value | 540B | -7% | 0% |
| 17 | DeepSeek V4 Pro deepseek-v4-pro |
DeepSeek | Flagship | $0.270 | $1.10 | 128K | 82 | 18 Low Value | 1.89T | +14% | -10% |
| 18 | Grok 4.1 Fast grok-4-1-fast |
xAI | Long-context | $0.200 | $1.00 | 256K | 72 | 18 Low Value | 754B | +13% | -20% |
| 19 | Grok Code Fast 1 grok-code-fast-1 |
xAI | Coding | $0.200 | $1.00 | 128K | 71 | 18 Low Value | 628B | +64% | 0% |
| 20 | Codestral 25.10 codestral-25-10 |
Mistral | Coding | $0.300 | $0.900 | 32K | 70 | 18 Low Value | 395B | +27% | 0% |
| 21 | MiniMax M3 minimax-m3 |
MiniMax | Reasoning | $0.300 | $1.20 | 1M | 80 | 16 Low Value | 3.81T | +158% | 0% |
| 22 | Grok 3 Mini grok-3-mini |
xAI | Budget | $0.200 | $1.00 | 64K | 62 | 16 Low Value | 222B | +9% | 0% |
| 23 | Kimi K2 Instruct kimi-k2-instruct |
Moonshot | Open | $0.300 | $1.20 | 200K | 72 | 15 Low Value | 744B | +66% | 0% |
| 24 | GPT-5.4 Mini gpt-5-4-mini |
OpenAI | Budget | $0.300 | $1.20 | 400K | 73 | 15 Low Value | 515B | +19% | -10% |
| 25 | MiMo Reasoner mimo-reasoner |
Xiaomi | Reasoning | $0.300 | $1.20 | 128K | 72 | 15 Low Value | 406B | +59% | 0% |
| 26 | MiMo Coder mimo-coder |
Xiaomi | Coding | $0.300 | $1.20 | 128K | 69 | 14 Low Value | 349B | +20% | 0% |
| 27 | Kimi K1.5 Long kimi-k1-5-long |
Moonshot | Long-context | $0.300 | $1.20 | 1M | 66 | 14 Low Value | 319B | +27% | 0% |
| 28 | Hunyuan-Large hunyuan-large |
Tencent | Open | $0.300 | $1.20 | 128K | 67 | 14 Low Value | 184B | +24% | 0% |
| 29 | DeepSeek V4.1 deepseek-v4-1 |
DeepSeek | Flagship | $0.420 | $1.68 | 512K | 85 | 12 Low Value | 2.37T | +117% | 0% |
| 30 | Kimi K2.6 kimi-k2-6 |
Moonshot | Coding | $0.400 | $1.60 | 256K | 76 | 12 Low Value | 988B | +28% | 0% |
| 31 | Kimi Coder kimi-coder |
Moonshot | Coding | $0.400 | $1.60 | 256K | 77 | 12 Low Value | 517B | +42% | 0% |
| 32 | Grok 4 Mini grok-4-mini |
xAI | Budget | $0.300 | $1.50 | 128K | 70 | 12 Low Value | 452B | +48% | 0% |
| 33 | Mixtral 8x22B mixtral-8x22b |
Mistral | Open | $0.900 | $0.900 | 64K | 64 | 12 Low Value | 142B | +9% | 0% |
| 34 | Hunyuan-Turbo hunyuan-turbo |
Tencent | Flagship | $0.400 | $1.60 | 128K | 74 | 11 Low Value | 396B | +46% | 0% |
| 35 | Devstral Medium devstral-medium |
Mistral | Coding | $0.500 | $1.50 | 128K | 73 | 11 Low Value | 301B | +52% | 0% |
| 36 | Hy3 Preview hy3-preview |
Tencent | Flagship | $0.500 | $2.00 | 256K | 79 | 10 Low Value | 3.42T | +8% | 0% |
| 37 | MiMo V2.5 Pro mimo-v2-5-pro |
Xiaomi | Flagship | $0.500 | $2.00 | 128K | 78 | 10 Low Value | 793B | +40% | 0% |
| 38 | Hunyuan-T1 hunyuan-t1 |
Tencent | Reasoning | $0.500 | $2.00 | 128K | 77 | 9 Low Value | 255B | +62% | 0% |
| 39 | DeepSeek R1 deepseek-r1 |
DeepSeek | Reasoning | $0.550 | $2.19 | 64K | 79 | 9 Low Value | 252B | -4% | 0% |
| 40 | Owl Alpha owl-alpha |
OpenRouter | Reasoning | $0.600 | $2.40 | 200K | 76 | 8 Low Value | 2.06T | -14% | 0% |
| 41 | Kimi K2 Thinking kimi-k2-thinking |
Moonshot | Reasoning | $0.600 | $2.50 | 256K | 81 | 8 Low Value | 929B | +86% | 0% |
| 42 | Kimi Researcher kimi-researcher |
Moonshot | Reasoning | $0.600 | $2.50 | 256K | 76 | 8 Low Value | 186B | +84% | 0% |
| 43 | Gemini 3.8 Flash gemini-3-8-flash |
Long-context | $0.750 | $3.75 | 1.0M | 87 | 6 Low Value | 113B | +1024% | 0% | |
| 44 | Claude Haiku 4.5 claude-haiku-4-5 |
Anthropic | Budget | $0.800 | $4.00 | 200K | 71 | 5 Low Value | 697B | +15% | 0% |
| 45 | Muse Spark 1.3 muse-spark-1-3 |
Meta | Flagship | $1.25 | $4.25 | 1.0M | 91 | 5 Low Value | 127B | +1008% | 0% |
| 46 | Grok 4.7 grok-4-7 |
xAI | Reasoning | $2.00 | $6.00 | 500K | 90 | 4 Low Value | 171B | +985% | 0% |
| 47 | Grok 4.6 grok-4-6 |
xAI | Reasoning | $2.00 | $6.00 | 2M | 88 | 3 Low Value | 1.41T | +113% | 0% |
| 48 | Magistral Medium 2 magistral-medium-2 |
Mistral | Reasoning | $2.00 | $5.00 | 128K | 73 | 3 Low Value | 471B | +88% | 0% |
| 49 | Pixtral Large pixtral-large |
Mistral | Flagship | $2.00 | $6.00 | 128K | 72 | 3 Low Value | 220B | +27% | 0% |
| 50 | Qwen3.8 Max qwen3-8-max |
Qwen | Flagship | $2.00 | $6.00 | 1M | 89 | 3 Low Value | 121B | +1016% | 0% |
| 51 | Gemini 3.1 Pro gemini-3-1-pro |
Long-context | $2.50 | $10.00 | 2M | 86 | 2 Low Value | 1.19T | -5% | -5% | |
| 52 | Mistral Large 3 mistral-large-3 |
Mistral | Flagship | $3.00 | $9.00 | 128K | 78 | 2 Low Value | 388B | -11% | 0% |
| 53 | Grok 3 Reasoning grok-3-reasoning |
xAI | Reasoning | $2.00 | $10.00 | 128K | 75 | 2 Low Value | 298B | -17% | 0% |
| 54 | Claude Sonnet 5.5 claude-sonnet-5-5 |
Anthropic | Flagship | $2.00 | $10.00 | 1M | 94 | 2 Low Value | 126B | +984% | 0% |
| 55 | MiMo V2.6 Pro UltraSpeed mimo-v2-6-pro-ultraspeed |
Xiaomi | Flagship | $4.35 | $8.70 | 1.0M | 90 | 2 Low Value | 126B | +982% | 0% |
| 56 | GPT-6 Sol gpt-6-sol |
OpenAI | Flagship | $2.00 | $10.00 | 1.1M | 90 | 2 Low Value | 119B | +1000% | 0% |
| 57 | GPT-6.1 Sol gpt-6-1-sol |
OpenAI | Flagship | $2.00 | $10.00 | 1.1M | 92 | 2 Low Value | 118B | +1003% | 0% |
| 58 | Claude Sonnet 4.6 claude-sonnet-4-6 |
Anthropic | Coding | $3.00 | $15.00 | 200K | 82 | 1 Low Value | 2.22T | +17% | 0% |
| 59 | GPT-5.6 gpt-5-6 |
OpenAI | Flagship | $4.00 | $20.00 | 1.1M | 91 | 1 Low Value | 1.81T | +50% | 0% |
| 60 | GPT-5.5 gpt-5-5 |
OpenAI | Flagship | $5.00 | $20.00 | 400K | 88 | 1 Low Value | 1.46T | +13% | -12% |
| 61 | Kimi K3 kimi-k3 |
Moonshot | Coding | $3.00 | $15.00 | 1M | 84 | 1 Low Value | 1.11T | +17% | 0% |
| 62 | Grok 4 Heavy grok-4-heavy |
xAI | Reasoning | $5.00 | $15.00 | 256K | 84 | 1 Low Value | 927B | +67% | 0% |
| 63 | Grok 4.1 grok-4.1 |
xAI | Long-context | $3.00 | $15.00 | 256K | 82 | 1 Low Value | 814B | +67% | 0% |
| 64 | GPT-5.4 gpt-5-4 |
OpenAI | Flagship | $5.00 | $20.00 | 400K | 85 | 1 Low Value | 774B | -18% | -10% |
| 65 | Grok 4 grok-4 |
xAI | Reasoning | $3.00 | $15.00 | 256K | 80 | 1 Low Value | 312B | -22% | 0% |
| 66 | Sonar Pro sonar-pro |
Perplexity | Long-context | $3.00 | $15.00 | 200K | 70 | 1 Low Value | 272B | -18% | 0% |
| 67 | Claude Opus 5.5 claude-opus-5-5 |
Anthropic | Reasoning | $4.00 | $20.00 | 1M | 95 | 1 Low Value | 124B | +1021% | 0% |
| 68 | Claude Fable 5 claude-fable-5 |
Anthropic | Flagship | $12.00 | $60.00 | 200K | 87 | 0 Low Value | 2.67T | +80% | -8% |
| 69 | GPT-6 Astra gpt-6-astra |
OpenAI | Flagship | $10.00 | $50.00 | 1.1M | 93 | 0 Low Value | 1.88T | +228% | 0% |
| 70 | Claude Opus 4.8 claude-opus-4-8 |
Anthropic | Flagship | $15.00 | $75.00 | 200K | 89 | 0 Low Value | 1.56T | +29% | 0% |
| 71 | Claude Opus 4.7 claude-opus-4-7 |
Anthropic | Flagship | $15.00 | $75.00 | 200K | 86 | 0 Low Value | 1.34T | -9% | 0% |
| 72 | Nemotron 3.5 Super VL nemotron-3-5-super-vl |
Nvidia | Open | $0.150 | $0.500 | 1.0M | — | — | 126B | +1014% | 0% |
| 73 | Gemini 4 Argon gemini-4-argon |
Flagship | $2.00 | $10.00 | 1.0M | — | — | 124B | +1003% | 0% | |
| 74 | GPT-6.1 Sol Pro gpt-6-1-sol-pro |
OpenAI | Reasoning | $2.00 | $10.00 | 1.1M | — | — | 123B | +1001% | 0% |
| 75 | GLM-5.3 FlashX glm-5-3-flashx |
Zhipu | Budget | $0.370 | $1.25 | 1.0M | — | — | 123B | +1007% | 0% |
| 76 | Command A+ command-a-plus |
Cohere | Flagship | $0.300 | $1.50 | 192K | — | — | 122B | +1009% | 0% |
| 77 | Muse Wonder muse-wonder |
Meta | Reasoning | $4.00 | $20.00 | 1.0M | — | — | 121B | +986% | 0% |
| 78 | Mistral Large 4 mistral-large-4 |
Mistral | Flagship | $0.680 | $2.09 | 1.0M | — | — | 121B | +978% | 0% |
| 79 | GPT-6 Luna Pro gpt-6-luna-pro |
OpenAI | Budget | $0.100 | $0.500 | 1.1M | — | — | 120B | +1015% | 0% |
| 80 | Qwen3.8 Omni Flash qwen3-8-omni-flash |
Qwen | Budget | $0.150 | $0.470 | 1M | — | — | 120B | +1004% | 0% |
| 81 | Muse Spark Open muse-spark-open |
Meta | Open | $1.25 | $4.25 | 1.0M | — | — | 119B | +993% | 0% |
| 82 | DeepSeek V4.1 Flash deepseek-v4-1-flash |
DeepSeek | Budget | $0.150 | $0.600 | 1.0M | — | — | 118B | +1019% | 0% |
| 83 | GLM-5.3 Prime glm-5-3-prime |
Zhipu | Reasoning | $2.80 | $8.80 | 1M | — | — | 118B | +988% | 0% |
| 84 | GPT-6 Sol Pro gpt-6-sol-pro |
OpenAI | Reasoning | $2.00 | $10.00 | 1.1M | — | — | 117B | +1019% | 0% |
| 85 | Claude Haiku 5.5 claude-haiku-5-5 |
Anthropic | Budget | $0.100 | $0.500 | 1M | — | — | 115B | +1017% | 0% |
| 86 | Qwen3.8 Max Prime qwen3-8-max-prime |
Qwen | Reasoning | $4.00 | $12.00 | 1M | — | — | 115B | +988% | 0% |
Token usage aggregated daily from Tokenscost calculator runs, public API mirrors, and recorded usage reports. Trailing 7-day window; figures refresh automatically each UTC day. Pricing coverage spans over 1,000 models.
Last updated: · Source: Pricing dataset · Methodology
Frequently asked questions
How are token costs calculated?
Multiply input tokens by the input rate and output tokens by the output rate, add the two amounts and divide by 1,000,000. A standard request example uses 1,000 input and 500 output tokens. Monthly budgets multiply the result by request volume; taxes, hidden reasoning tokens, tools and tier surcharges can add costs.
Which recorded model has the lowest request cost?
LFM 2.2 6b by Liquid AI costs $0.00002 for the standard request, with $0.01 input and $0.02 output per million tokens. This price ranking does not establish quality equivalence; verify task success and context requirements.
Can cached and batch discounts be combined?
Only recorded rates are used. Cached input can have cache-write or storage charges, while batch rates apply only to eligible asynchronous jobs. We do not automatically stack the discounts or treat an absent rate as zero. Check the linked provider source before budgeting.
How current is the public dataset?
The live pricing table synchronizes stored model records daily; public file snapshots update when the site is rebuilt. Synchronization is not independent provider verification. It contains 1,026 priced models across 124 providers, last updated 2026-10-31. Directory entries are counted separately. A synchronization date is not proof of independent verification.