Capability ranking
Best AI for Instruction Following
Best Instruction-Following Models
Updated 2026-08-11
Ranked models235
Vendors25
Open-weight models51
Official prices105
What this ranking says
A quick reading of the current instruction ranking before the full data table.
GLM 5.2 leads this page at 100.0, 1.8 points ahead of GPT-5.2-Pro.
GLM 5.2 is the highest-ranked open-weight option at #1. Nova Micro has the lowest official input price among ranked models.
Instruction Ranking Leaders
The leading instruction scores, with the axis focused on the competitive range rather than zero.
Instruction Capability Profile
Compare instruction strength with each model's broader capability score.
ModelReasoningKnowledgeMathCodingInstruction
Instruction scoreOverall score
80 models with comparable dataOpen a model by selecting its logo
Instruction Model Ranking
Models are ordered by their instruction score.
| 1 | ZA | GLM-5.2 | 100.0 | 79.3 | 92.7 | N/A | 89.5 | 56.5 | 80.0% | 1M | Open weights | $1.4 | $4.4 |
| 2 | OP | GPT-5.2 Pro | 98.2 | 71.6 | 78.5 | 66.7 | 79.9 | 63.2 | 100.0% | 400K | Closed weights | $21 | $168 |
| 3 | XA | Grok-4 Heavy | 98.2 | 68.2 | 88.2 | N/A | 76.8 | 86.1 | 80.0% | N/A | Closed weights | N/A | N/A |
| 4 | AC | Qwen3.8 Max | 97.9 | 89.3 | 97.8 | N/A | 70.7 | 85.8 | 80.0% | 1M | Closed weights | $2 | $6 |
| 5 | MA | Kimi K3 | 97.9 | 93.0 | 97.0 | 50.0 | 95.7 | 94.1 | 100.0% | 1M | Open weights | $3 | $15 |
| 6 | ME | LongCat-Flash-Thinking-2601 | 93.4 | 60.8 | 63.5 | N/A | 28.6 | 76.2 | 80.0% | 128K | Closed weights | N/A | N/A |
| 7 | AN | Claude Mythos Preview | 93.3 | 93.6 | 99.8 | 100.0 | 99.7 | 92.7 | 100.0% | N/A | Closed weights | N/A | N/A |
| 8 | BY | Seed 2.1 Pro | 93.1 | 78.7 | 80.1 | 91.7 | 58.8 | 64.3 | 100.0% | N/A | Closed weights | N/A | N/A |
| 9 | OP | GPT-5.6 Sol | 90.6 | 95.1 | 87.8 | 95.8 | 100.0 | 92.9 | 100.0% | 1.1M | Closed weights | $5 | $30 |
| 10 | AC | Qwen3.7-Plus | 90.1 | 73.8 | 72.1 | 92.0 | 62.4 | 59.0 | 100.0% | 1M | Closed weights | $0.50 | $3 |
| 11 | NV | Nemotron 3 Ultra (550B A55B) | 89.9 | 62.9 | 82.6 | N/A | 53.6 | 46.0 | 80.0% | 1M | Open weights | $0.50 | $2.5 |
| 12 | ME | Muse Spark | 89.8 | 73.2 | 78.9 | 100.0 | 95.7 | 49.4 | 100.0% | N/A | Closed weights | N/A | N/A |
| 13 | AC | Qwen3.7 Max | 89.6 | 78.8 | 95.7 | N/A | 77.4 | 83.1 | 80.0% | 1M | Closed weights | $2.5 | $7.5 |
| 14 | AN | Claude Opus 4.7 | 89.5 | 84.2 | 98.7 | N/A | 88.6 | 69.3 | 80.0% | 1M | Closed weights | $5 | $25 |
| 15 | BY | Seed 2.0 Pro | 89.4 | 69.5 | 88.8 | N/A | N/A | 75.9 | 60.0% | 256K | Closed weights | N/A | N/A |
| 16 | ME | Muse Spark 1.1 | 89.4 | 88.3 | 87.5 | N/A | 96.7 | 75.7 | 80.0% | 1M | Closed weights | $1.25 | $4.25 |
| 17 | AN | Claude Fable 5 | 89.2 | 95.8 | N/A | 100.0 | 97.8 | 94.6 | 80.0% | 1M | Closed weights | $10 | $50 |
| 18 | ST | Step-3.5-Flash | 88.5 | 60.4 | N/A | N/A | 55.6 | 57.8 | 60.0% | 256K | Open weights | $0.10 | $0.30 |
| 19 | BY | Seed 2.1 Turbo | 87.5 | 73.1 | 52.6 | 41.7 | 66.6 | 53.5 | 100.0% | N/A | Closed weights | N/A | N/A |
| 20 | AN | Claude Sonnet 5 | 87.2 | 84.0 | N/A | 62.5 | 71.7 | 78.8 | 80.0% | 1M | Closed weights | $2 | $10 |
| 21 | OP | GPT-5.2 | 85.6 | 71.4 | 75.5 | 60.7 | 76.1 | 59.1 | 100.0% | 400K | Closed weights | $1.75 | $14 |
| 22 | DE | DeepSeek-V4-Pro-Max | 85.4 | 73.8 | 90.1 | N/A | 86.0 | 63.2 | 80.0% | 1M | Closed weights | $0.435 | $0.87 |
| 23 | AN | Claude Opus 4.8 | 84.7 | 87.9 | 99.0 | 37.5 | 94.6 | 85.0 | 100.0% | 1M | Closed weights | $5 | $25 |
| 24 | ZA | GLM-4.5 | 84.5 | 49.3 | 61.4 | N/A | 56.8 | 58.4 | 80.0% | 131.1K | Open weights | $0.60 | $2.2 |
| 25 | DE | DeepSeek-V3.2-Speciale | 83.2 | 58.1 | N/A | N/A | 74.0 | 46.1 | 60.0% | 131.1K | Closed weights | $0.14 | $0.28 |
| 26 | XA | Grok 4 Fast | 82.1 | 54.0 | 79.4 | N/A | 50.5 | 53.2 | 80.0% | 2M | Closed weights | N/A | N/A |
| 27 | ZA | GLM-4.7 | 81.4 | 59.7 | 79.4 | N/A | 53.7 | 47.9 | 80.0% | 204.8K | Open weights | $0.60 | $2.2 |
| 28 | AC | Qwen2-VL-72B-Instruct | 81.3 | 25.5 | 57.1 | 34.4 | 18.2 | N/A | 80.0% | N/A | Closed weights | N/A | N/A |
| 29 | AN | Claude 3.5 Sonnet | 80.4 | 32.1 | 80.4 | 45.2 | 76.1 | 48.8 | 100.0% | 200K | Closed weights | N/A | N/A |
| 30 | AC | Qwen3.5-397B-A17B | 80.0 | 66.3 | 90.8 | N/A | 53.7 | 69.8 | 80.0% | 262.1K | Open weights | $0.60 | $3.6 |
| 31 | AC | Qwen3-235B-A22B-Thinking-2507 | 80.0 | 48.5 | 71.4 | N/A | 46.6 | 40.4 | 80.0% | 262.1K | Closed weights | N/A | N/A |
| 32 | ME | Llama 4 Maverick | 79.5 | 24.9 | 53.8 | 61.3 | 58.6 | 43.2 | 100.0% | 1M | Closed weights | N/A | N/A |
| 33 | AN | Claude Opus 4.5 | 78.9 | 69.5 | 61.3 | 100.0 | N/A | 71.1 | 80.0% | 200K | Closed weights | $5 | $25 |
| 34 | BA | ERNIE 5.0 | 78.8 | 58.7 | 77.3 | N/A | 62.0 | N/A | 60.0% | N/A | Closed weights | N/A | N/A |
| 35 | OP | GPT-5.1 | 77.9 | 62.1 | 86.5 | 96.8 | 50.0 | 70.9 | 100.0% | 400K | Closed weights | $1.25 | $10 |
| 36 | OP | GPT-5.1 Instant | 77.9 | 61.3 | 86.5 | 96.8 | 50.0 | 70.9 | 100.0% | 400K | Closed weights | N/A | N/A |
| 37 | AC | Qwen3.6 Plus | 77.3 | 67.5 | 84.9 | 82.0 | 74.4 | 48.8 | 100.0% | 1M | Closed weights | $0.50 | $3 |
| 38 | AN | Claude Opus 4.6 | 77.2 | 79.6 | 77.5 | 50.0 | 85.9 | 72.2 | 100.0% | 1M | Closed weights | $5 | $25 |
| 39 | ST | Step3-VL-10B | 77.2 | 43.6 | N/A | 75.8 | 72.2 | N/A | 60.0% | N/A | Closed weights | N/A | N/A |
| 40 | DE | DeepSeek-R1-0528 | 77.2 | 44.2 | 65.2 | N/A | 28.3 | 22.5 | 80.0% | 131.1K | Closed weights | N/A | N/A |
| 41 | ZA | GLM-4.6 | 76.1 | 54.2 | 65.2 | N/A | 29.9 | 50.4 | 80.0% | 204.8K | Open weights | $0.60 | $2.2 |
| 42 | GO | Gemma 4 31B | 76.0 | 60.1 | 69.5 | 63.6 | 62.1 | 71.7 | 100.0% | 262.1K | Open weights | N/A | N/A |
| 43 | XA | Grok-3 | 75.2 | 54.8 | 76.2 | 74.2 | N/A | 86.1 | 80.0% | 128K | Closed weights | N/A | N/A |
| 44 | MI | MAI-Thinking-1 | 75.1 | 55.1 | 74.5 | 13.6 | 30.0 | 54.4 | 100.0% | N/A | Closed weights | N/A | N/A |
| 45 | AM Nova ProAmazon | Nova Pro | 75.0 | 20.9 | 65.3 | 66.1 | 80.7 | 52.7 | 100.0% | 300K | Closed weights | $0.80 | $3.2 |
| 46 | ZA | GLM-4.5-Air | 74.9 | 41.5 | 54.3 | N/A | 51.1 | 48.2 | 80.0% | 131.1K | Open weights | $0.20 | $1.1 |
| 47 | OP | GPT-4 | 74.9 | 4.6 | 8.6 | N/A | 36.3 | 10.8 | 80.0% | 8.2K | Closed weights | $30 | $60 |
| 48 | GO | Gemini 3.1 Pro | 74.1 | 77.5 | 91.8 | N/A | 82.6 | 55.9 | 80.0% | 1M | Closed weights | $2 | $12 |
| 49 | AMNova 2 LiteAmazon | Nova 2 Lite | 73.6 | 43.0 | 35.7 | N/A | N/A | 35.5 | 60.0% | 1M | Closed weights | $0.33 | $2.75 |
| 50 | GO | Gemini 1.5 Pro | 72.8 | 20.5 | 31.3 | 40.3 | 52.0 | 41.5 | 100.0% | 2.1M | Closed weights | N/A | N/A |
| 51 | XI | MiMo-V2.5-Pro | 72.7 | 46.7 | 70.0 | 88.2 | 82.7 | 50.1 | 100.0% | 1M | Open weights | $0.435 | $0.87 |
| 52 | AN | Claude 3 Opus | 72.7 | 16.7 | 59.2 | N/A | 60.5 | 46.1 | 80.0% | 200K | Closed weights | N/A | N/A |
| 53 | AC | Qwen3.6-27B | 72.7 | 60.5 | 79.4 | 83.8 | 54.8 | 45.5 | 100.0% | 262.1K | Open weights | $0.60 | $3.6 |
| 54 | BY | Seed 2.0 Lite | 72.6 | 53.6 | 77.7 | N/A | N/A | 65.6 | 60.0% | N/A | Closed weights | N/A | N/A |
| 55 | GO | Gemini 2.5 Pro Preview 06-05 | 72.6 | 49.9 | 81.1 | 70.7 | 39.1 | 66.1 | 100.0% | 1M | Closed weights | $1.25 | $10 |
| 56 | AN | Claude Sonnet 4.6 | 72.3 | 72.6 | 70.1 | N/A | 78.3 | 62.2 | 80.0% | 1M | Closed weights | $3 | $15 |
| 57 | OP | GPT-5.5 | 72.2 | 84.0 | 98.8 | 100.0 | 73.1 | 73.8 | 100.0% | 1.1M | Closed weights | $5 | $30 |
| 58 | AC | Qwen3.6-35B-A3B | 72.1 | 55.3 | 57.0 | 74.0 | 47.4 | 26.5 | 100.0% | 262.1K | Open weights | $0.248 | $1.49 |
| 59 | SA | Sarvam-30B | 71.8 | 34.3 | 39.1 | N/A | 30.5 | 49.6 | 80.0% | 128K | Open weights | N/A | N/A |
| 60 | MA | Kimi K2.6 | 71.5 | 75.5 | 78.5 | N/A | 69.0 | 70.5 | 80.0% | 262.1K | Open weights | $0.95 | $4 |
| 61 | OP | GPT-5.4 | 71.2 | 77.1 | 90.3 | 83.3 | 72.2 | 57.8 | 100.0% | 1.1M | Closed weights | $2.5 | $15 |
| 62 | AMNova 2 ProAmazon | Nova 2 Pro | 70.2 | 52.5 | 71.4 | N/A | N/A | 57.3 | 60.0% | N/A | Closed weights | N/A | N/A |
| 63 | AC | Qwen3.5-27B | 69.3 | 57.2 | 59.4 | 68.6 | 78.5 | 44.5 | 100.0% | 262.1K | Open weights | $0.30 | $2.4 |
| 64 | OP | GPT-4 Turbo | 69.2 | 14.6 | 21.5 | N/A | 70.3 | 55.4 | 80.0% | 128K | Closed weights | $10 | $30 |
| 65 | NR | Hermes 3 70B | 68.8 | 8.0 | 29.7 | 35.7 | 34.0 | N/A | 80.0% | 131.1K | Closed weights | N/A | N/A |
| 66 | ME | Llama 3.3 70B Instruct | 68.7 | 22.0 | 24.9 | N/A | 68.6 | 66.9 | 80.0% | 128K | Open weights | N/A | N/A |
| 67 | AC | Qwen3.5-122B-A10B | 68.5 | 59.2 | 61.7 | 87.8 | 73.4 | 51.6 | 100.0% | 262.1K | Open weights | $0.40 | $3.2 |
| 68 | ME | Llama 3.1 405B Instruct | 68.0 | 24.5 | 61.1 | N/A | 76.5 | 87.7 | 80.0% | 128K | Closed weights | N/A | N/A |
| 69 | MA | Kimi K2-Thinking-0905 | 67.0 | 61.0 | 75.5 | N/A | 60.2 | 55.0 | 80.0% | 262.1K | Closed weights | $0.60 | $2.5 |
| 70 | OP | GPT-5.6 Terra | 66.0 | 87.6 | 71.4 | 50.0 | 71.9 | 79.1 | 100.0% | 1.1M | Closed weights | $2 | $12 |
| 71 | AC | Qwen3-235B-A22B-Instruct-2507 | 65.8 | 42.4 | 61.1 | 0.0 | 20.8 | 40.2 | 100.0% | 262.1K | Open weights | $0.70 | $2.8 |
| 72 | MA | Pixtral Large | 65.6 | 21.2 | 82.1 | 38.7 | 63.2 | N/A | 80.0% | 128K | Closed weights | N/A | N/A |
| 73 | XA | Grok-4 | 65.5 | 56.5 | 51.9 | N/A | 47.8 | 83.3 | 80.0% | 256K | Closed weights | N/A | N/A |
| 74 | NV | Nemotron 3 Nano (30B A3B) | 65.2 | 35.2 | 54.3 | N/A | 25.0 | 9.7 | 80.0% | 262.1K | Open weights | N/A | N/A |
| 75 | SA | Sarvam-105B | 65.1 | 44.6 | 60.1 | N/A | 49.9 | 28.8 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 76 | ZA | GLM-4.7-Flash | 64.6 | 38.5 | 55.8 | N/A | 20.1 | 17.7 | 80.0% | 200K | Open weights | N/A | N/A |
| 77 | NV | Nemotron 3 Super (120B A12B) | 64.5 | 44.8 | 70.0 | N/A | 60.8 | 23.5 | 80.0% | 262.1K | Open weights | $0.20 | $0.80 |
| 78 | AC | Qwen3-Next-80B-A3B-Thinking | 64.5 | 40.9 | 58.2 | N/A | 37.5 | 41.2 | 80.0% | 131.1K | Open weights | $0.50 | $6 |
| 79 | XI | MiMo-V2.5 | 64.3 | 61.8 | 33.9 | N/A | N/A | 43.8 | 60.0% | 1M | Open weights | $0.14 | $0.28 |
| 80 | GO | Gemini 3 Pro | 64.3 | 67.3 | 68.8 | 100.0 | 36.4 | 60.5 | 100.0% | 1M | Closed weights | N/A | N/A |
| 81 | GO | Gemma 2 27B | 64.2 | 0.1 | 69.3 | 97.1 | 25.6 | 8.6 | 100.0% | N/A | Closed weights | N/A | N/A |
| 82 | OP | o3-mini | 63.9 | 36.3 | 58.1 | N/A | 56.3 | 36.2 | 80.0% | 200K | Closed weights | $1.1 | $4.4 |
| 83 | GO | Gemma 4 26B-A4B | 63.8 | 52.6 | 57.3 | 54.5 | 47.2 | 67.5 | 100.0% | 262.1K | Open weights | N/A | N/A |
| 84 | OP | GPT-5 mini | 63.3 | 44.9 | 68.5 | N/A | 33.4 | N/A | 60.0% | 400K | Closed weights | $0.25 | $2 |
| 85 | AN | Claude Sonnet 4.5 | 62.9 | 56.0 | 72.1 | 33.3 | N/A | 90.8 | 80.0% | 200K | Closed weights | $3 | $15 |
| 86 | ME | Muse Glimmer-30B | 62.9 | 58.5 | 60.1 | N/A | 40.8 | 45.3 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 87 | MI | MiniMax M3 | 62.6 | 71.3 | 69.2 | 74.0 | 100.0 | 53.8 | 100.0% | 512K | Open weights | $0.30 | $1.2 |
| 88 | AC | Qwen3 VL 235B A22B Instruct | 62.5 | 43.2 | 65.5 | 45.3 | 44.6 | 59.3 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 89 | AC | Qwen3 VL 235B A22B Thinking | 62.2 | 45.8 | 53.9 | 45.3 | 48.7 | 61.6 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 90 | OP | o3 | 62.1 | 52.6 | 50.1 | 63.8 | 47.3 | 51.0 | 100.0% | 200K | Closed weights | $2 | $8 |
| 91 | MA | Kimi-k1.5 | 62.0 | 30.1 | N/A | 49.2 | 70.3 | N/A | 60.0% | N/A | Closed weights | N/A | N/A |
| 92 | DE | DeepSeek-V3.1 | 61.6 | 39.9 | 53.6 | N/A | 14.6 | 31.4 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 93 | DE | DeepSeek R1 Distill Qwen 7B | 61.5 | 15.7 | 23.2 | N/A | 25.8 | 29.9 | 80.0% | N/A | Closed weights | N/A | N/A |
| 94 | XA | Grok-3 Mini | 61.1 | 48.4 | 73.6 | N/A | N/A | 90.3 | 60.0% | 128K | Closed weights | N/A | N/A |
| 95 | GO | Gemini 2.0 Flash | 61.0 | 28.1 | 33.7 | 54.8 | 97.2 | 27.8 | 100.0% | 1M | Closed weights | N/A | N/A |
| 96 | DE | DeepSeek-V2.5 | 60.9 | 15.2 | N/A | N/A | 69.4 | 38.2 | 60.0% | 8.2K | Closed weights | N/A | N/A |
| 97 | GO | Gemini 3.5 Flash | 60.5 | 79.2 | 90.0 | N/A | 65.2 | 64.5 | 80.0% | 1M | Closed weights | $1.5 | $9 |
| 98 | MA | Kimi K2.5 | 60.3 | 66.0 | 69.0 | 46.0 | 73.6 | 48.1 | 100.0% | 262.1K | Open weights | $0.60 | $3 |
| 99 | ME | Llama 4 Scout | 60.0 | 14.1 | 56.3 | 46.8 | 44.3 | 23.8 | 100.0% | 10M | Closed weights | N/A | N/A |
| 100 | XI | MiMo-V2-Flash | 60.0 | 51.3 | 73.0 | N/A | 38.4 | 44.9 | 80.0% | 262.1K | Open weights | $0.14 | $0.28 |
| 101 | AMNova 2 OmniAmazon | Nova 2 Omni | 59.4 | 43.9 | 20.9 | N/A | N/A | 48.0 | 60.0% | N/A | Closed weights | N/A | N/A |
| 102 | OP | o1-mini | 59.1 | 16.8 | 32.6 | N/A | 12.9 | 90.8 | 80.0% | 128K | Closed weights | N/A | N/A |
| 103 | MI | MiniMax M2.1 | 58.9 | 53.7 | 65.2 | N/A | 40.8 | 51.5 | 80.0% | 204.8K | Open weights | $0.30 | $1.2 |
| 104 | AC | Qwen3.5-35B-A3B | 58.2 | 52.0 | 39.3 | 52.4 | 59.2 | 42.7 | 100.0% | 262.1K | Open weights | $0.25 | $2 |
| 105 | CO | Command R+ | 58.2 | 0.0 | 42.4 | N/A | 10.6 | N/A | 60.0% | 128K | Closed weights | $0.15 | $0.60 |
| 106 | AC | Qwen3 VL 32B Thinking | 58.1 | 40.5 | 65.5 | 64.8 | 60.6 | 60.2 | 100.0% | N/A | Closed weights | N/A | N/A |
| 107 | GO | Gemini 3 Flash | 57.9 | 64.8 | 71.1 | 96.0 | 69.6 | 46.1 | 100.0% | 1M | Closed weights | $0.50 | $3 |
| 108 | AC | Qwen3 Max | 57.6 | 40.7 | 33.0 | N/A | N/A | 38.0 | 60.0% | 262.1K | Closed weights | $1.2 | $6 |
| 109 | MI | Phi-3.5-MoE-instruct | 57.6 | 4.9 | 60.9 | 100.0 | 34.3 | 39.7 | 100.0% | N/A | Closed weights | N/A | N/A |
| 110 | AN | Claude Opus 4.1 | 57.4 | 52.4 | 64.2 | 66.7 | N/A | 75.8 | 80.0% | 200K | Closed weights | N/A | N/A |
| 111 | IN | Mercury 2 | 56.6 | 39.3 | 51.7 | N/A | N/A | 35.2 | 60.0% | 128K | Closed weights | $0.25 | $0.75 |
| 112 | OP | o1 | 56.3 | 36.3 | 59.4 | 72.6 | 66.3 | 46.7 | 100.0% | 200K | Closed weights | $15 | $60 |
| 113 | OP | GPT OSS 120B High | 55.9 | 44.0 | 64.2 | N/A | N/A | 47.9 | 60.0% | 131.1K | Closed weights | N/A | N/A |
| 114 | MI | MAI-Code-1-Flash | 55.7 | 50.4 | 76.2 | N/A | 20.0 | 31.1 | 80.0% | 256K | Closed weights | N/A | N/A |
| 115 | ME | LongCat-Flash-Chat | 55.5 | 36.9 | 50.2 | N/A | 54.8 | 42.8 | 80.0% | 128K | Closed weights | N/A | N/A |
| 116 | OP | GPT-4o mini | 55.3 | 8.1 | 12.0 | 29.0 | 42.9 | 28.5 | 100.0% | 128K | Closed weights | $0.15 | $0.60 |
| 117 | AC | Qwen2.5 72B Instruct | 54.9 | 20.9 | 22.5 | N/A | 80.1 | 60.3 | 80.0% | 131.1K | Closed weights | $1.4 | $5.6 |
| 118 | ZA | GLM-5.1 | 54.8 | 69.6 | 80.7 | N/A | 68.0 | 47.1 | 80.0% | 200K | Open weights | $1.4 | $4.4 |
| 119 | MA | Mistral Large 2 | 54.5 | 7.1 | N/A | N/A | 63.8 | 86.9 | 60.0% | 128K | Closed weights | N/A | N/A |
| 120 | OP | GPT-5.4 mini | 54.4 | 60.5 | 72.5 | N/A | 51.1 | 37.7 | 80.0% | 400K | Closed weights | $0.75 | $4.5 |
| 121 | AC | Qwen3-Next-80B-A3B-Instruct | 54.1 | 36.6 | 53.4 | N/A | 16.7 | 28.4 | 80.0% | 131.1K | Open weights | $0.50 | $2 |
| 122 | AC | Qwen2.5 32B Instruct | 53.6 | 14.8 | 30.0 | N/A | 81.1 | 68.3 | 80.0% | N/A | Closed weights | $0.70 | $2.8 |
| 123 | AN | Claude 3 Sonnet | 53.6 | 3.5 | 48.8 | N/A | 38.7 | 20.0 | 80.0% | 200K | Closed weights | N/A | N/A |
| 124 | NV | Nemotron Nano 9B v2 | 52.9 | 32.4 | 34.3 | N/A | 87.1 | 75.0 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 125 | OP | GPT-4.5 | 52.8 | 36.2 | 43.8 | 69.3 | 85.0 | 40.5 | 100.0% | 128K | Closed weights | N/A | N/A |
| 126 | GO | Gemini 2.5 Flash | 52.5 | 38.1 | 70.6 | 79.0 | 14.1 | 46.1 | 100.0% | 1M | Closed weights | $0.30 | $2.5 |
| 127 | MI | MiniMax M2 | 52.4 | 43.9 | 59.4 | N/A | 17.4 | 41.4 | 80.0% | 196.6K | Open weights | $0.30 | $1.2 |
| 128 | AC | Qwen2 72B Instruct | 51.9 | 9.8 | 22.8 | N/A | 61.0 | 46.6 | 80.0% | N/A | Closed weights | N/A | N/A |
| 129 | GO | Gemma 3 12B | 51.9 | 11.2 | 14.1 | 10.0 | 57.3 | 32.4 | 100.0% | 131.1K | Closed weights | N/A | N/A |
| 130 | CO | Command A+ | 51.8 | 41.9 | 33.1 | 67.7 | 86.8 | 33.3 | 100.0% | 128K | Open weights | $2.5 | $10 |
| 131 | AN | Claude 3.7 Sonnet | 51.3 | 38.7 | 76.8 | 66.1 | 56.5 | 60.5 | 100.0% | 200K | Closed weights | N/A | N/A |
| 132 | AN | Claude Opus 4 | 50.3 | 46.4 | 37.7 | 33.3 | N/A | 68.6 | 80.0% | 200K | Closed weights | N/A | N/A |
| 133 | GO | Gemma 3 27B | 50.1 | 15.8 | 20.4 | 20.0 | 69.1 | 39.2 | 100.0% | 131.1K | Closed weights | N/A | N/A |
| 134 | GO | Gemini 3.1 Flash-Lite | 50.1 | 51.9 | 71.8 | 80.0 | 27.2 | 10.3 | 100.0% | 1M | Closed weights | $0.25 | $1.5 |
| 135 | MA | Kimi K2-Instruct-0905 | 50.0 | 37.2 | 55.1 | N/A | 42.5 | 31.7 | 80.0% | N/A | Closed weights | $0.60 | $2.5 |
| 136 | IB | Granite 3.3 8B Instruct | 49.7 | 3.8 | N/A | N/A | 13.4 | 80.8 | 60.0% | 128K | Closed weights | N/A | N/A |
| 137 | MA | Mistral Medium 3.5 | 48.6 | 49.2 | N/A | N/A | 25.0 | 38.3 | 60.0% | 262.1K | Open weights | $1.5 | $7.5 |
| 138 | NV | Llama 3.1 Nemotron 70B Instruct | 48.5 | 3.1 | 33.3 | N/A | 53.2 | N/A | 60.0% | 128K | Open weights | N/A | N/A |
| 139 | MI | MiniMax M1 40K | 48.3 | 35.3 | 43.4 | N/A | 29.6 | 50.9 | 80.0% | N/A | Closed weights | N/A | N/A |
| 140 | NV | Llama 3.1 Nemotron Ultra 253B v1 | 48.1 | 30.5 | 57.1 | N/A | 75.8 | 66.7 | 80.0% | N/A | Closed weights | N/A | N/A |
| 141 | NV | Llama-3.3 Nemotron Super 49B v1 | 48.0 | 22.4 | 39.5 | N/A | 64.5 | 96.9 | 80.0% | 131.1K | Open weights | N/A | N/A |
| 142 | OP | GPT-5 nano | 47.8 | 31.1 | 47.2 | N/A | 16.8 | N/A | 60.0% | 400K | Closed weights | $0.05 | $0.40 |
| 143 | GO | Gemini 1.5 Flash | 46.6 | 10.7 | 26.0 | 33.9 | 40.4 | 23.1 | 100.0% | 1M | Closed weights | N/A | N/A |
| 144 | AC | Qwen3.5-9B | 46.6 | 41.4 | 58.9 | N/A | 28.1 | 33.5 | 80.0% | 262.1K | Open weights | N/A | N/A |
| 145 | MI | MiniMax M1 80K | 46.3 | 37.0 | 45.1 | N/A | 39.0 | 56.9 | 80.0% | 1M | Closed weights | N/A | N/A |
| 146 | DE | DeepSeek R1 Distill Llama 8B | 46.1 | 12.4 | 22.5 | N/A | 9.7 | 31.9 | 80.0% | N/A | Closed weights | N/A | N/A |
| 147 | XA | Grok-2 | 45.3 | 21.0 | 29.2 | 41.9 | 62.4 | 66.9 | 100.0% | 128K | Closed weights | N/A | N/A |
| 148 | AN | Claude 3 Haiku | 44.5 | 0.0 | 37.0 | N/A | 29.2 | 27.7 | 80.0% | 200K | Closed weights | N/A | N/A |
| 149 | OP | o4-mini | 44.5 | 45.0 | 66.7 | 82.3 | 57.2 | 52.2 | 100.0% | 200K | Closed weights | $1.1 | $4.4 |
| 150 | AC | Qwen3 VL 32B Instruct | 43.9 | 35.2 | 58.8 | 70.6 | 48.9 | 46.4 | 100.0% | N/A | Closed weights | N/A | N/A |
| 151 | AMNova LiteAmazon | Nova Lite | 42.6 | 13.2 | 51.1 | 35.5 | 59.9 | 34.2 | 100.0% | 300K | Closed weights | $0.06 | $0.24 |
| 152 | GO | Gemma 4 12B | 41.9 | 38.2 | 51.8 | 36.4 | 32.8 | 29.3 | 100.0% | N/A | Closed weights | N/A | N/A |
| 153 | ME | Llama 3.1 70B Instruct | 41.9 | 13.2 | 51.8 | N/A | 77.6 | 61.2 | 80.0% | 128K | Closed weights | N/A | N/A |
| 154 | DE | DeepSeek-V3 0324 | 41.9 | 26.0 | 42.1 | N/A | 32.3 | 36.1 | 80.0% | 163.8K | Closed weights | N/A | N/A |
| 155 | MI | Phi 4 Reasoning Plus | 41.5 | 26.8 | 42.7 | N/A | N/A | 66.3 | 60.0% | N/A | Closed weights | $0.125 | $0.50 |
| 156 | GO | Gemma 2 9B | 41.3 | 0.0 | 48.0 | 72.3 | 18.4 | 2.3 | 100.0% | N/A | Closed weights | N/A | N/A |
| 157 | ME | Llama 3.2 90B Instruct | 41.3 | 10.8 | 35.0 | 30.6 | 32.5 | N/A | 80.0% | 128K | Closed weights | N/A | N/A |
| 158 | OP | GPT-5.6 Luna | 40.8 | 76.4 | 55.2 | 25.0 | 43.8 | 68.6 | 100.0% | 1.1M | Closed weights | $0.20 | $1.2 |
| 159 | GO | DiffusionGemma 26B-A4B | 40.5 | 34.2 | 33.4 | 45.5 | 34.0 | 18.6 | 100.0% | N/A | Closed weights | N/A | N/A |
| 160 | GO | Gemini 1.0 Pro | 40.2 | 0.0 | 3.4 | 4.8 | 4.1 | N/A | 80.0% | 32.8K | Closed weights | N/A | N/A |
| 161 | AC | QwQ-32B | 40.0 | 24.8 | 36.5 | N/A | 21.0 | 34.9 | 80.0% | N/A | Closed weights | $0.287 | $0.861 |
| 162 | XA | Grok-2 mini | 39.8 | 16.3 | 26.0 | 37.1 | 56.1 | 50.8 | 100.0% | N/A | Closed weights | N/A | N/A |
| 163 | GO | Gemini 2.0 Flash-Lite | 39.7 | 21.0 | 26.6 | 43.5 | 86.0 | N/A | 80.0% | 1M | Closed weights | N/A | N/A |
| 164 | AC | Qwen3 VL 30B A3B Thinking | 39.6 | 31.8 | 45.6 | 52.3 | 43.3 | 38.8 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 165 | OP | GPT-4.1 | 39.0 | 31.9 | 38.6 | 64.5 | 26.7 | 37.7 | 100.0% | 1M | Closed weights | $2 | $8 |
| 166 | AC | Qwen3 30B A3B | 38.7 | 32.1 | 37.5 | N/A | N/A | 49.2 | 60.0% | 128K | Closed weights | N/A | N/A |
| 167 | AN | Claude Sonnet 4 | 38.6 | 39.5 | 56.2 | 62.9 | N/A | 65.9 | 80.0% | 200K | Closed weights | N/A | N/A |
| 168 | OP | GPT-5.5 Instant | 38.4 | 53.7 | 67.7 | 6.3 | N/A | N/A | 60.0% | 400K | Closed weights | N/A | N/A |
| 169 | AC | Qwen3 VL 8B Thinking | 38.3 | 28.5 | 39.3 | 30.8 | 38.2 | 28.8 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 170 | GO | Gemma 3n E4B Instructed | 38.2 | 0.0 | 8.0 | N/A | 37.5 | 16.9 | 80.0% | 32K | Closed weights | N/A | N/A |
| 171 | AN | Claude 3.5 Haiku | 38.0 | 9.3 | 13.7 | N/A | 56.8 | 19.9 | 80.0% | 200K | Closed weights | N/A | N/A |
| 172 | MI | Phi-3.5-vision-instruct | 37.5 | 0.0 | 18.7 | 3.2 | 2.6 | N/A | 80.0% | N/A | Closed weights | N/A | N/A |
| 173 | ME | LongCat-Flash-Lite | 37.5 | 33.0 | 40.3 | N/A | 69.3 | 29.6 | 80.0% | 256K | Closed weights | N/A | N/A |
| 174 | AC | Qwen2.5 VL 72B Instruct | 37.2 | 22.7 | 42.5 | 44.6 | 20.8 | 12.6 | 100.0% | 131.1K | Open weights | N/A | N/A |
| 175 | MA | Ministral 3 (8B Reasoning 2512) | 37.2 | 29.0 | 40.8 | N/A | N/A | 55.6 | 60.0% | 262.1K | Closed weights | $0.10 | $0.10 |
| 176 | AC | Qwen3 VL 30B A3B Instruct | 36.7 | 28.8 | 31.7 | 44.0 | 28.7 | 26.8 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 177 | MA | Mistral Small 3.2 24B Instruct | 36.4 | 15.9 | 57.7 | 35.5 | 45.8 | N/A | 80.0% | N/A | Closed weights | N/A | N/A |
| 178 | MA | Mistral Large 3 (675B Instruct 2512) | 36.4 | 15.8 | 16.9 | N/A | 70.0 | 24.3 | 80.0% | 262.1K | Open weights | $0.50 | $1.5 |
| 179 | GO | Gemini 2.5 Flash-Lite | 36.1 | 19.1 | 35.2 | 59.7 | 3.3 | 9.8 | 100.0% | 1M | Closed weights | $0.10 | $0.40 |
| 180 | MA | Mistral Small 4 | 36.0 | 33.4 | 35.9 | N/A | N/A | 73.9 | 60.0% | 256K | Open weights | $0.15 | $0.60 |
| 181 | OP | GPT-5.4 nano | 35.6 | 49.1 | 54.5 | N/A | 45.6 | 19.8 | 80.0% | 400K | Closed weights | $0.20 | $1.25 |
| 182 | AC | Qwen3.5-4B | 35.5 | 32.4 | 45.4 | N/A | 18.8 | 20.6 | 80.0% | N/A | Closed weights | N/A | N/A |
| 183 | AC | Qwen2.5-Coder 32B Instruct | 33.5 | 4.4 | 39.4 | N/A | 46.2 | 68.0 | 80.0% | 128K | Closed weights | $0.287 | $0.861 |
| 184 | AN | Claude Haiku 4.5 | 33.0 | 45.5 | 48.9 | 0.0 | N/A | 55.1 | 80.0% | 200K | Closed weights | $1 | $5 |
| 185 | AC | Qwen2.5 14B Instruct | 32.7 | 8.2 | 21.4 | N/A | 62.1 | 51.2 | 80.0% | N/A | Closed weights | $0.35 | $1.4 |
| 186 | GO | Gemma 4 E4B | 31.7 | 24.4 | 22.2 | 18.2 | 25.8 | 18.2 | 100.0% | 131.1K | Open weights | N/A | N/A |
| 187 | MI | Phi-3.5-mini-instruct | 31.5 | 0.0 | 50.7 | 62.5 | 24.4 | 18.7 | 100.0% | 128K | Closed weights | N/A | N/A |
| 188 | OP | GPT-4.1 mini | 30.9 | 24.8 | 35.6 | 58.1 | 28.4 | 13.1 | 100.0% | 1M | Closed weights | $0.40 | $1.6 |
| 189 | GO | Gemini 2.0 Flash Thinking | 30.8 | 29.9 | 52.6 | 71.0 | N/A | N/A | 60.0% | N/A | Closed weights | N/A | N/A |
| 190 | MI | Phi 4 Reasoning | 30.5 | 22.0 | 37.5 | N/A | N/A | 73.6 | 60.0% | N/A | Closed weights | $0.125 | $0.50 |
| 191 | MA | Mistral Small 3 24B Instruct | 29.1 | 7.5 | 18.0 | N/A | 48.6 | 43.9 | 80.0% | 32K | Closed weights | N/A | N/A |
| 192 | GO | Gemma 3n E4B Instructed LiteRT Preview | 29.0 | 0.0 | 21.1 | 41.2 | 31.7 | 16.9 | 100.0% | N/A | Closed weights | N/A | N/A |
| 193 | AL | Jamba 1.5 Large | 27.4 | 2.7 | 46.1 | N/A | 31.9 | N/A | 60.0% | 256K | Closed weights | N/A | N/A |
| 194 | AC | Qwen3 VL 8B Instruct | 27.1 | 24.8 | 39.2 | 16.0 | 21.0 | 29.3 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 195 | MA | Mistral NeMo Instruct | 26.5 | 0.0 | 83.3 | 33.1 | N/A | N/A | 60.0% | 128K | Closed weights | $0.15 | $0.15 |
| 196 | GO | Gemma 3 4B | 26.2 | 0.0 | 1.7 | 0.0 | 38.9 | 10.8 | 100.0% | 131.1K | Closed weights | N/A | N/A |
| 197 | MA | Min istral 3 (3B Reasoning 2512) | 26.1 | 20.8 | 27.9 | N/A | N/A | 48.6 | 60.0% | 131.1K | Closed weights | $0.04 | $0.04 |
| 198 | ME | Llama 3.2 11B Instruct | 26.1 | 0.0 | 27.3 | 8.1 | 16.8 | N/A | 80.0% | 128K | Closed weights | N/A | N/A |
| 199 | MA | Ministral 8B Instruct | 26.1 | 0.0 | 48.5 | 17.6 | 19.1 | 0.0 | 100.0% | 131.1K | Open weights | N/A | N/A |
| 200 | AC | Qwen3 VL 4B Thinking | 25.8 | 23.0 | 27.2 | 28.3 | 27.5 | 25.5 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 201 | MI | Phi 4 Mini | 25.1 | 0.0 | 45.0 | 62.5 | 35.6 | N/A | 80.0% | 128K | Open weights | $0.075 | $0.30 |
| 202 | AC | Qwen2.5-Omni-7B | 25.1 | 10.2 | 21.8 | 27.7 | 35.3 | 26.1 | 100.0% | N/A | Closed weights | $0.10 | $0.40 |
| 203 | MI | Phi-4-multimodal-instruct | 25.0 | 7.4 | 25.6 | 17.7 | 36.8 | N/A | 80.0% | 128K | Closed weights | N/A | N/A |
| 204 | AC | Qwen2.5 7B Instruct | 24.6 | 6.5 | 9.4 | N/A | 58.4 | 37.6 | 80.0% | 131.1K | Closed weights | $0.175 | $0.70 |
| 205 | OP | GPT-4o | 23.7 | 23.9 | 42.9 | 30.2 | 19.8 | 10.8 | 100.0% | 128K | Closed weights | $2.5 | $10 |
| 206 | GO | Gemini Diffusion | 23.2 | 3.6 | 12.7 | N/A | N/A | 36.8 | 60.0% | N/A | Closed weights | N/A | N/A |
| 207 | MA | Pixtral-12B | 20.0 | 0.0 | 28.3 | 11.3 | 23.8 | 16.9 | 100.0% | 128K | Open weights | $0.15 | $0.15 |
| 208 | GO | Gemma 4 E2B | 19.9 | 11.6 | 11.0 | 9.1 | 19.4 | 8.1 | 100.0% | 131.1K | Open weights | N/A | N/A |
| 209 | AC | Qwen3 VL 4B Instruct | 19.6 | 19.3 | 20.8 | 12.1 | 12.9 | 17.8 | 100.0% | 262.1K | Closed weights | N/A | N/A |
| 210 | MA | Mistral Small 3.1 24B Instruct | 19.6 | 8.4 | 18.9 | 53.0 | 40.7 | 59.2 | 100.0% | N/A | Closed weights | N/A | N/A |
| 211 | MA | Magistral Medium | 18.6 | 19.8 | 46.4 | N/A | 10.9 | 29.7 | 80.0% | N/A | Closed weights | N/A | N/A |
| 212 | MI | Phi 4 | 18.1 | 9.4 | 29.6 | N/A | 56.1 | 38.5 | 80.0% | 16K | Closed weights | $0.125 | $0.50 |
| 213 | AC | Qwen2 7B Instruct | 17.3 | 0.0 | 3.0 | N/A | 14.9 | 17.1 | 80.0% | N/A | Closed weights | N/A | N/A |
| 214 | AC | Qwen2.5 VL 7B Instruct | 16.9 | 12.8 | 37.6 | 23.8 | 8.4 | 6.7 | 100.0% | N/A | Closed weights | $0.35 | $1.05 |
| 215 | ZA | GLM-5V-Turbo | 16.7 | 57.0 | 0.0 | N/A | N/A | 80.0 | 60.0% | 200K | Closed weights | $5 | $22 |
| 216 | AC | Qwen2.5-Coder 7B Instruct | 16.0 | 0.0 | 12.1 | N/A | 19.9 | 50.5 | 80.0% | N/A | Closed weights | $0.144 | $0.287 |
| 217 | MA | Magistral Small 2506 | 15.9 | 19.8 | 41.6 | N/A | N/A | 41.7 | 60.0% | 131.1K | Open weights | N/A | N/A |
| 218 | AC | Qwen2.5 VL 32B Instruct | 15.2 | 17.9 | 32.0 | 49.2 | 38.6 | 49.4 | 100.0% | N/A | Closed weights | N/A | N/A |
| 219 | NV | Llama 3.1 Nemotron Nano 8B V1 | 14.7 | 10.4 | 28.8 | N/A | 45.2 | 84.4 | 80.0% | N/A | Closed weights | N/A | N/A |
| 220 | OP | GPT-3.5 Turbo | 13.6 | 0.0 | 5.2 | 0.0 | 13.9 | 12.3 | 100.0% | 16.4K | Closed weights | $0.50 | $1.5 |
| 221 | XA | Grok-1.5 | 13.3 | 0.0 | 9.0 | 13.7 | 26.9 | 21.5 | 100.0% | N/A | Closed weights | N/A | N/A |
| 222 | IB | IBM Granite 4.0 Tiny Preview | 13.2 | 0.0 | N/A | N/A | 7.7 | 36.9 | 60.0% | N/A | Closed weights | N/A | N/A |
| 223 | GO | Gemini 1.5 Flash 8B | 13.1 | 0.0 | 11.2 | 16.1 | 24.5 | N/A | 80.0% | 1M | Closed weights | N/A | N/A |
| 224 | AMNova MicroAmazon | Nova Micro | 13.0 | 4.3 | 40.6 | 0.0 | 38.3 | 17.3 | 100.0% | 128K | Closed weights | $0.035 | $0.14 |
| 225 | GO | Gemma 3n E2B Instructed LiteRT (Preview) | 12.5 | 0.0 | 5.2 | 15.1 | 12.3 | 7.0 | 100.0% | N/A | Closed weights | N/A | N/A |
| 226 | GO | Gemma 3n E2B Instructed | 12.4 | 0.0 | 1.8 | N/A | 12.5 | 7.0 | 80.0% | N/A | Closed weights | N/A | N/A |
| 227 | AC | Qwen3.5-2B | 11.0 | 2.8 | 18.5 | N/A | N/A | 7.6 | 60.0% | N/A | Closed weights | N/A | N/A |
| 228 | ME | Llama 3.2 3B Instruct | 10.7 | 0.0 | 29.1 | N/A | 16.7 | N/A | 60.0% | 128K | Closed weights | N/A | N/A |
| 229 | ME | Llama 3.1 8B Instruct | 10.2 | 0.0 | 29.3 | N/A | 10.3 | 49.2 | 80.0% | 131.1K | Closed weights | N/A | N/A |
| 230 | AL | Jamba 1.5 Mini | 7.5 | 0.0 | 34.8 | N/A | 14.9 | N/A | 60.0% | 256.1K | Closed weights | N/A | N/A |
| 231 | OP | GPT-4.1 nano | 6.1 | 4.3 | 24.0 | 19.4 | 21.1 | 0.0 | 100.0% | 1M | Closed weights | $0.10 | $0.40 |
| 232 | GO | Gemma 3 1B | 2.5 | 0.0 | 0.4 | N/A | 6.9 | 1.0 | 80.0% | N/A | Closed weights | N/A | N/A |
| 233 | AC | Qwen3.5-0.8B | 1.2 | 0.0 | 3.3 | N/A | N/A | 0.0 | 60.0% | N/A | Closed weights | N/A | N/A |
| 234 | BA | ERNIE 4.5 | 0.1 | 0.0 | 25.9 | 0.0 | 0.0 | 0.0 | 100.0% | 128K | Closed weights | N/A | N/A |
| 235 | DE | DeepSeek VL2 Tiny | 0.0 | 0.0 | 16.6 | 1.6 | 15.8 | N/A | 80.0% | N/A | Closed weights | N/A | N/A |
Official PAYG prices appear here. Third-party offers remain on the pricing and model detail pages.
Capability at each price point
Official vendor API prices plotted against the instruction metric used on this page. Price remains a separate decision signal.
Instruction scoreOfficial API price blend (8:1 input/output), USD / 1M
Efficient frontier105 models with official PAYG prices
Instruction Ranking Composition
Vendor concentration and model access among the first 25 ranked products.
Top-model vendor mix
25 modelsCompare the leaders
Open focused comparisons between the current leader and the nearest practical alternatives.
About this leaderboard
Common questions about Best AI for Instruction Following.
What is the best model on Best AI for Instruction Following?
GLM 5.2 is currently ranked first with a instruction score of 100.0.
How is this leaderboard ranked?
Each model appears once using its current scored version. Category pages use the capability named in the title, while the overall page uses the LLMBoard score.
How many models are included?
This page currently ranks 235 unique model products.
Where does pricing come from?
Main price columns use the model vendor's official standard PAYG API rate. Eligible third-party offers appear only in separately labeled columns, and unavailable official prices display as N/A.
Does Arena affect the score?
No. Arena results are displayed as an independent signal and are not included in the current LLMBoard capability score.