IFBench benchmark
322 AI models ranked on IFBench. Each row shows the model’s score next to its cheapest API price per 1M tokens where it has one, so you can weigh quality against cost.
Top model today
Grok 4.3 (xAI) leads at 83%, from $1.25 per 1M input via xAI.
Data via arXiv:2507.02833 →Models ranked
322
on this benchmark
Top score
83%
current leader
Type
Single eval
benchmark
Updated
Oct 14, 2026
benchmark scores
Cost vs IFBench
209 priced models = reasoning model (extended thinking)
Each labelled name links to that model’s page. Curated to the cost-efficiency frontier, top scorers and cheapest so labels stay readable; the full field is in the leaderboard table below (and via “select all”). Reasoning models (extended “thinking” before answering) carry a ring around the dot.
Models ranked on IFBench
highest score first · 322 modelsMaker
| # | ||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Grok 4.3 |  xAI | 83%medium | $1.25 via xAI | $2.50 via xAI | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 2 | Grok 4 20 0309 | xAI | 83% | $1.25 via xAI | $2.50 via xAI | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 3 | MiniMax M3 |  MiniMax | 83% | $0.23 via Wandb | $0.96 via Wandb | |||||||||||||||||||||||||
| 4 | Nvidia Nemotron 3 Ultra 550B A55b |  NVIDIA | 81% | $0.50 via DeepInfra | $2.15 via Wandb | |||||||||||||||||||||||||
| 5 | Grok 4.20 | xAI | 81% | $1.25 via xAI | $2.50 via xAI | |||||||||||||||||||||||||
| 6 | Qwen3.7 Max |  Alibaba | 81% | $1.25 via Novita | $3.75 via Novita | |||||||||||||||||||||||||
| 7 | Nemotron Cascade 2 30B A3B | NVIDIA | 80% | n/a | n/a | |||||||||||||||||||||||||
| 8 | MiMo-V2.5-Pro |  Xiaomi | 80% | $0.435 Routed quote | $0.87 Routed quote | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 9 | Nova 2.0 Pro Preview |  Amazon | 80%low | n/a | n/a | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 10 | DeepSeek V4 Flash 0423 |  DeepSeek | 79% | $0.09 via DeepInfra | $0.18 via DeepInfra | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 11 | Qwen3.5 397B A17B | Alibaba | 79% | $0.164 via Aihubmix | $0.986 via Aihubmix | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 12 | Qwen3.7 Plus | Alibaba | 78% | $0.282 via Aihubmix | $1.13 via Aihubmix | |||||||||||||||||||||||||
| 13 | Gemini 3 Flash |  Google | 78%reasoning: true | $0.50 Routed quote | $3.00 Routed quote | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 14 | GPT-5.2-Codex |  OpenAI | 78% | $1.75 via Databricks | $14.00 via Databricks | |||||||||||||||||||||||||
| 15 | Gemini 3.1 Flash Lite Preview | Google | 77% | $0.25 via Google | $1.50 via Google | |||||||||||||||||||||||||
| 16 | Gemini 3.1 Pro Preview | Google | 77% | $2.00 via Google | $12.00 via Google | |||||||||||||||||||||||||
| 17 | Qwen3.6 Max Preview | Alibaba | 77% | n/a | n/a | |||||||||||||||||||||||||
| 18 | DeepSeek V4 Pro 0423 | DeepSeek | 76% | $0.435 via Tencent | $0.87 via Tencent | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 19 | Gemini 3.5 Flash | Google | 76% | $1.50 via Google | $9.00 via Google | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 20 | GLM 5.1 |  Zhipu | 76% | $1.05 via DeepInfra | $3.50 via DeepInfra | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 21 | Kimi K2.6 |  Moonshot | 76% | $0.65 via Wandb | $3.40 via Novita | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 22 | GPT-5.4 Nano | OpenAI | 76% | $0.20 via Databricks | $1.25 via Databricks | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 23 | Muse Spark |  Meta | 76% | n/a | n/a | |||||||||||||||||||||||||
| 24 | GPT-5.5 | OpenAI | 76% | $5.00 via OpenAI | $30.00 via OpenAI | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 25 | MiniMax M2.7 | MiniMax | 76% | $0.25 via DeepInfra | $0.55 via Pinstripes | |||||||||||||||||||||||||
| 26 | Qwen3.5-122B-A10B | Alibaba | 76% | $0.113 via Aihubmix | $0.901 via Aihubmix | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 27 | Qwen3.5-27B | Alibaba | 76% | $0.26 via DeepInfra | $2.40 via Novita | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 28 | Gemma 4 31B | Google | 76% | $0.14 via Amazon Bedrock | $0.40 via Amazon Bedrock | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 29 | GPT-5 Mini | OpenAI | 75% | $0.25 via Databricks | $1.20 via Snowflake | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 30 | GPT-5.2 | OpenAI | 75% | $1.75 via OpenAI | $14.00 via OpenAI | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 31 | GPT-5.3-Codex | OpenAI | 75% | $1.75 via Aihubmix | $14.00 via Aihubmix | |||||||||||||||||||||||||
| 32 | Qwen3.6 Plus | Alibaba | 75% | $0.325 Routed quote | $1.95 Routed quote | |||||||||||||||||||||||||
| 33 | GPT 5 Codex | OpenAI | 74% | $1.25 Routed quote | $10.00 Routed quote | |||||||||||||||||||||||||
| 34 | GPT-5.4 | OpenAI | 74% | $2.50 via Databricks | $15.00 via OpenAI | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 35 | Command A+ |  Cohere | 74% | $0.30 Routed quote | $1.50 Routed quote | |||||||||||||||||||||||||
| 36 | Gemma 4 12B | Google | 74% | n/a | n/a | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 37 | GLM 5.2 | Zhipu | 73% | $0.05 Routed quote | $1.98 via Scx AI | |||||||||||||||||||||||||
| 38 | GPT-5.4 Mini | OpenAI | 73% | $0.75 via Databricks | $4.50 via OpenAI | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 39 | GLM 5 Turbo | Zhipu | 73% | $1.20 via Novita | $4.00 via Novita | |||||||||||||||||||||||||
| 40 | GPT-5 | OpenAI | 73% | $1.25 via Databricks | $10.00 via Databricks | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 41 | GPT-5.1 | OpenAI | 73% | $1.25 via Databricks | $10.00 via Databricks | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 42 | GPT-5.6 Sol | OpenAI | 73% | $2.00 Routed quote | $10.00 Routed quote | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 43 | Qwen3.5-35B-A3B | Alibaba | 73% | $0.14 via DeepInfra | $1.00 via DeepInfra | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 44 | Gemma 4 26B A4b | Google | 72% | $0.13 via Amazon Bedrock | $0.40 via Amazon Bedrock | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 45 | GLM 5 | Zhipu | 72% | $0.60 via DeepInfra | $1.92 Routed quote | |||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||
| 46 | MiniMax M2 | MiniMax | 72% | $0.255 Routed quote | $1.02 Routed quote | |||||||||||||||||||||||||
| 47 | MiMo-V2-Flash (Feb 2026) (0206) | Xiaomi | 72% | n/a | n/a | |||||||||||||||||||||||||
| 48 | MiniMax M2.5 | MiniMax | 72% | $0.27 Routed quote | $1.08 Routed quote | |||||||||||||||||||||||||
| 49 | Nvidia Nemotron 3 Super 120B A12b | NVIDIA | 71% | $0.085 via DeepInfra | $0.40 via DeepInfra | |||||||||||||||||||||||||
| 50 | GPT-5.5 Instant (May 2026) | OpenAI | 71% | n/a | n/a | |||||||||||||||||||||||||
1–50 of 323
Page 1 of 7
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments can be materially lower. Cost per task uses best-effort token usage. An OpenRouter routed quote is the price of whichever upstream provider OpenRouter prefers today, so it moves day to day.
Frequently asked questions
What is the IFBench benchmark?
IFBench (Ai2) tests how well models follow new, previously unseen verifiable output constraints such as word-count limits, required keywords or formatting rules that were not part of training. Scoring is fully programmatic: verifier functions check whether each response satisfies the constraint.
Which AI model scores highest on IFBench?
Grok 4.3 (xAI) leads with 83%, from $1.25 per 1M input tokens from xAI.
How many models are ranked on IFBench?
322 models carry an IFBench score. Scores come from Artificial Analysis; where a model has a listed API price it is drawn from LiteLLM and OpenRouter, and models with no current price are listed without one.
Other benchmarks
compare the same models on a different evalIntelligence Indexcomposite indexCoding Indexcomposite indexAgentic Indexcomposite indexGPQA DiamondleaderboardHumanity's Last ExamleaderboardMMLU-ProleaderboardMMMU-ProleaderboardLiveCodeBenchleaderboardSciCodeleaderboardTerminalBench HardleaderboardTerminalBench v2.1leaderboardτ²-benchleaderboardτ-bench BankingleaderboardAPEX AgentsleaderboardIT-Bench SREleaderboardGDPvalleaderboardCritPtleaderboardOmniscienceleaderboardLong-Context Reasoningcomposite indexSWE-bench VerifiedleaderboardSWE-bench ProleaderboardAIME 2026leaderboardARC-AGI-2leaderboardAA-Briefcaseleaderboard
Benchmark scores via Artificial Analysis · scores updated Oct 14, 2026 · Listed prices from LiteLLM, OpenRouter and Vercel · catalogue refreshed Oct 14, 2026 · 131 prices older, back to Jul 19, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI