Recently available
Trending Local Models
- Loading the Hugging Face attention signal…
This week's release radar
Language model rankings
95 current modelsDeterministic lower-confidence-bound capability ordering from public data. Current and All versions are separate contiguous positions; v5 posterior rank analytics do not apply until a validated cutover.
95 shown
1Current-model rank | Anthropic | 92.3 4 public benchmark observations across 4 boards; evidence state Measured; 87.7% support; score interval 92.3–98.3. | 1M 128K max output Exact context: 1,000,000 tokens; exact maximum output: 128,000 tokens | 1,507 Arena Highest supported configuration: Default; source rank 1 | $50.00 |
2Current-model rank | Anthropic | 90.9 5 public benchmark observations across 5 boards; evidence state Measured; 88.8% support; score interval 90.9–96.7. | 1M 128K max output Exact context: 1,000,000 tokens; exact maximum output: 128,000 tokens | 1,493 Arena Highest supported configuration: High; source rank 8 | $25.00 |
3Current-model rank | Qwen3.8 MaxNew Qwen | 88.9 2 public benchmark observations across 2 boards; evidence state Measured; 56.0% support; score interval 88.9–99.9. | 1M 131K max output Exact context: 1,000,000 tokens; exact maximum output: 131,072 tokens | 1,497 Arena Highest supported configuration: Default; source rank 6 | $6.00 |
4Current-model rank | 88.7 4 public benchmark observations across 4 boards; evidence state Measured; 83.5% support; score interval 88.7–95.3. | 1.0M Text + Image Exact context: 1,048,576 tokens | 1,485 Arena Highest supported configuration: Max; source rank 14 | $15.00 | |
5Current-model rank | OpenAI | 88.4 5 public benchmark observations across 5 boards; evidence state Measured; 88.1% support; score interval 88.4–94.4. | 1.1M 128K max output Exact context: 1,050,000 tokens; exact maximum output: 128,000 tokens | 1,482 Arena Highest supported configuration: X-High; source rank 18 | $30.00 |
6Current-model rank | OpenAI | 87.4 5 public benchmark observations across 5 boards; evidence state Measured; 90.3% support; score interval 87.4–93.0. | 1.1M 128K max output Exact context: 1,050,000 tokens; exact maximum output: 128,000 tokens | 1,482 Arena Highest supported configuration: High; source rank 17 | $30.00 |
7Current-model rank | Meta | 87.4 2 public benchmark observations across 2 boards; evidence state Measured; 37.8% support; score interval 87.4–100.0. | 1.0M Text + Image + Video + File + Audio Exact context: 1,048,576 tokens | 1,498 Arena Highest supported configuration: X-High; source rank 4 | $4.25 |
8Current-model rank | Google | 86.9 3 public benchmark observations across 3 boards; evidence state Measured; 87.5% support; score interval 86.9–92.9. | 1.0M 66K max output Exact context: 1,048,576 tokens; exact maximum output: 65,536 tokens | 1,487 Arena Highest supported configuration: Default; source rank 12 | $12.00 |
9Current-model rank | Google | 85.4 4 public benchmark observations across 4 boards; evidence state Measured; 85.4% support; score interval 85.4–91.8. | 1.0M 66K max output Exact context: 1,048,576 tokens; exact maximum output: 65,536 tokens | 1,485 Arena Highest supported configuration: Default; source rank 15 | $7.50 |
10Current-model rank | 84.9 2 public benchmark observations across 2 boards; evidence state Measured; 82.5% support; score interval 84.9–91.7. | 2M Text + Image + File Exact context: 2,000,000 tokens | 1,471 Arena Highest supported configuration: Beta · 0309; source rank 32 | $2.50 | |
11Current-model rank | xAI | 83.6 5 public benchmark observations across 5 boards; evidence state Measured; 88.9% support; score interval 83.6–89.4. | 500K Text + Image + File Exact context: 500,000 tokens | 1,468 Arena Highest supported configuration: Default; source rank 35 | $6.00 |
12Current-model rank | OpenAI | 83.4 5 public benchmark observations across 5 boards; evidence state Measured; 88.7% support; score interval 83.4–89.2. | 1.1M 128K max output Exact context: 1,050,000 tokens; exact maximum output: 128,000 tokens | 1,467 Arena Highest supported configuration: X-High; source rank 38 | $6.00 |
13Current-model rank | Anthropic | 83.4 3 public benchmark observations across 3 boards; evidence state Measured; 85.8% support; score interval 83.4–89.6. | 1M 128K max output Exact context: 1,000,000 tokens; exact maximum output: 128,000 tokens | 1,463 Arena Highest supported configuration: High; source rank 43 | $10.00 |
14Current-model rank | 83.3 4 public benchmark observations across 4 boards; evidence state Measured; 88.7% support; score interval 83.3–89.1. | 1.0M 131K max output Exact context: 1,048,576 tokens; exact maximum output: 131,072 tokens | 1,471 Arena Highest supported configuration: Max; source rank 33 | $1.58 | |
15Current-model rank | XiaomiOpen | 81.4 3 public benchmark observations across 3 boards; evidence state Measured; 87.2% support; score interval 81.4–87.4. | 1.1M 131K max output Exact context: 1,050,000 tokens; exact maximum output: 131,072 tokens | 1,467 Arena Highest supported configuration: Default; source rank 39 | $0.87 |
16Current-model rank | DeepSeekOpen | 79.3 3 public benchmark observations across 3 boards; evidence state Measured; 87.4% support; score interval 79.3–85.3. | 1.0M 384K max output Exact context: 1,048,576 tokens; exact maximum output: 384,000 tokens | 1,457 Arena Highest supported configuration: Default; source rank 50 | $0.87 |
17Current-model rank | Qwen | 79.1 3 public benchmark observations across 3 boards; evidence state Measured; 86.8% support; score interval 79.1–85.3. | 1M 131K max output Exact context: 1,000,000 tokens; exact maximum output: 131,072 tokens | 1,458 Arena Highest supported configuration: Default; source rank 49 | $1.28 |
18Current-model rank | OpenAI | 78.7 2 public benchmark observations across 2 boards; evidence state Measured; 82.0% support; score interval 78.7–85.5. | 128K 16K max output Exact context: 128,000 tokens; exact maximum output: 16,384 tokens | 1,449 Arena Highest supported configuration: Default; source rank 63 | $14.00 |
19Current-model rank | OpenAI | 78.1 5 public benchmark observations across 5 boards; evidence state Measured; 88.6% support; score interval 78.1–83.9. | 1.1M 128K max output Exact context: 1,050,000 tokens; exact maximum output: 128,000 tokens | 1,450 Arena Highest supported configuration: X-High; source rank 60 | $0.60 |
20Current-model rank | Google | 77.7 4 public benchmark observations across 4 boards; evidence state Measured; 85.2% support; score interval 77.7–84.1. | 1.0M 66K max output Exact context: 1,048,576 tokens; exact maximum output: 65,536 tokens | 1,459 Arena Highest supported configuration: Default; source rank 47 | $2.50 |
21Current-model rank | OpenAI | 77.6 3 public benchmark observations across 3 boards; evidence state Measured; 86.3% support; score interval 77.6–83.8. | 400K 128K max output Exact context: 400,000 tokens; exact maximum output: 128,000 tokens | 1,448 Arena Highest supported configuration: High; source rank 64 | $4.50 |
22Current-model rank | 76.9 2 public benchmark observations across 2 boards; evidence state Measured; 84.2% support; score interval 76.9–83.5. | 262K 66K max output Exact context: 262,144 tokens; exact maximum output: 65,536 tokens | 1,442 Arena Highest supported configuration: Default; source rank 73 | $2.34 | |
23Current-model rank | MiniMaxOpen | 75.2 3 public benchmark observations across 3 boards; evidence state Measured; 87.1% support; score interval 75.2–81.2. | 1.0M 512K max output Exact context: 1,048,576 tokens; exact maximum output: 512,000 tokens | 1,445 Arena Highest supported configuration: Default; source rank 70 | $1.20 |
24Current-model rank | 75.1 3 public benchmark observations across 3 boards; evidence state Measured; 55.0% support; score interval 75.1–86.3. | 262K 128K max output Exact context: 262,144 tokens; exact maximum output: 128,000 tokens | 1,453 Arena Highest supported configuration: Default; source rank 57 | $0.528 | |
25Current-model rank | 74.7 2 public benchmark observations across 2 boards; evidence state Measured; 84.0% support; score interval 74.7–81.3. | 1.1M 131K max output Exact context: 1,050,000 tokens; exact maximum output: 131,072 tokens | 1,434 Arena Highest supported configuration: Default; source rank 85 | $0.28 | |
26Current-model rank | GoogleOpen | 74.0 3 public benchmark observations across 3 boards; evidence state Measured; 66.4% support; score interval 74.0–83.4. | 262K 262K max output Exact context: 262,144 tokens; exact maximum output: 262,144 tokens | 1,451 Arena Highest supported configuration: Default; source rank 58 | $0.34 |
27Current-model rank | GoogleOpen | 73.7 2 public benchmark observations across 2 boards; evidence state Measured; 63.3% support; score interval 73.7–83.5. | 262K 16K max output Exact context: 262,144 tokens; exact maximum output: 16,384 tokens | 1,438 Arena Highest supported configuration: Default; source rank 78 | $0.34 |
28Current-model rank | Z.ai | 73.6 2 public benchmark observations across 2 boards; evidence state Measured; 76.1% support; score interval 73.6–81.4. | 203K 131K max output Exact context: 202,752 tokens; exact maximum output: 131,072 tokens | 1,433 Arena Highest supported configuration: Default; source rank 87 | $4.00 |
29Current-model rank | 73.3 4 public benchmark observations across 4 boards; evidence state Measured; 87.2% support; score interval 73.3–79.3. | 1.0M Text + Image + Audio Exact context: 1,048,576 tokens | 1,442 Arena Highest supported configuration: Default; source rank 76 | $4.05 | |
30Current-model rank | Mistral AI | 72.0 2 public benchmark observations across 2 boards; evidence state Measured; 78.5% support; score interval 72.0–79.4. | 262K Text + Image + File Exact context: 262,144 tokens | 1,427 Arena Highest supported configuration: Default; source rank 94 | $7.50 |
31Current-model rank | 71.3 2 public benchmark observations across 2 boards; evidence state Measured; 76.2% support; score interval 71.3–79.1. | 512K Text Exact context: 512,288 tokens | 1,426 Arena Highest supported configuration: Default; source rank 96 | $2.40 | |
32Current-model rank | Moonshot AIOpen | 70.8 3 public benchmark observations across 3 boards; evidence state Measured; 86.4% support; score interval 70.8–77.0. | 262K 100K max output Exact context: 262,144 tokens; exact maximum output: 100,352 tokens | 1,430 Arena Highest supported configuration: Turbo; source rank 93 | $2.50 |
33Current-model rank | OpenAI | 70.6 3 public benchmark observations across 3 boards; evidence state Measured; 84.6% support; score interval 70.6–77.0. | 200K 100K max output Exact context: 200,000 tokens; exact maximum output: 100,000 tokens | 1,431 Arena Highest supported configuration: Default; source rank 91 | $8.00 |
34Current-model rank | 70.5 3 public benchmark observations across 3 boards; evidence state Measured; 85.6% support; score interval 70.5–76.9. | 262K 16K max output Exact context: 262,144 tokens; exact maximum output: 16,384 tokens | 1,423 Arena Highest supported configuration: Default; source rank 103 | $0.55 | |
35Current-model rank | DeepSeekOpen | 70.0 4 public benchmark observations across 4 boards; evidence state Measured; 90.9% support; score interval 70.0–75.4. | 164K 66K max output Exact context: 163,840 tokens; exact maximum output: 65,536 tokens | 1,425 Arena Highest supported configuration: Default; source rank 98 | $0.40 |
36Current-model rank | 69.4 2 public benchmark observations across 2 boards; evidence state Measured; 81.4% support; score interval 69.4–76.4. | 164K 33K max output Exact context: 163,840 tokens; exact maximum output: 32,768 tokens | 1,422 Arena Highest supported configuration: Default; source rank 106 | $2.15 | |
37Current-model rank | 68.0 2 public benchmark observations across 2 boards; evidence state Measured; 83.6% support; score interval 68.0–74.6. | 262K 82K max output Exact context: 262,144 tokens; exact maximum output: 81,920 tokens | 1,417 Arena Highest supported configuration: Default; source rank 113 | $2.40 | |
38Current-model rank | Mistral AI | 67.8 2 public benchmark observations across 2 boards; evidence state Measured; 82.5% support; score interval 67.8–74.6. | 262K Text + Image + File Exact context: 262,144 tokens | 1,415 Arena Highest supported configuration: Default; source rank 117 | $1.50 |
39Current-model rank | 67.1 2 public benchmark observations across 2 boards; evidence state Measured; 79.0% support; score interval 67.1–74.5. | 262K 33K max output Exact context: 262,144 tokens; exact maximum output: 32,768 tokens | 1,415 Arena Highest supported configuration: Default; source rank 118 | $1.90 | |
40Current-model rank | Anthropic | 66.0 3 public benchmark observations across 3 boards; evidence state Measured; 84.6% support; score interval 66.0–72.4. | 200K 64K max output Exact context: 200,000 tokens; exact maximum output: 64,000 tokens | 1,412 Arena Highest supported configuration: Default; source rank 122 | $5.00 |
41Current-model rank | DeepSeekOpen | 64.5 2 public benchmark observations across 2 boards; evidence state Measured; 50.5% support; score interval 64.5–76.5. | 164K 33K max output Exact context: 163,840 tokens; exact maximum output: 32,768 tokens | 1,417 Arena Highest supported configuration: Thinking; source rank 112 | $1.00 |
42Current-model rank | OpenAI | 63.5 3 public benchmark observations across 3 boards; evidence state Measured; 86.3% support; score interval 63.5–69.7. | 400K 128K max output Exact context: 400,000 tokens; exact maximum output: 128,000 tokens | 1,403 Arena Highest supported configuration: High; source rank 133 | $1.25 |
43Current-model rank | 63.3 2 public benchmark observations across 2 boards; evidence state Measured; 82.9% support; score interval 63.3–70.1. | 262K 16K max output Exact context: 262,144 tokens; exact maximum output: 16,384 tokens | 1,401 Arena Highest supported configuration: Default; source rank 135 | $1.10 | |
44Current-model rank | 62.1 2 public benchmark observations across 2 boards; evidence state Measured; 76.3% support; score interval 62.1–69.9. | 262K Text Exact context: 262,144 tokens | 1,399 Arena Highest supported configuration: Default; source rank 138 | $2.30 | |
45Current-model rank | 61.7 2 public benchmark observations across 2 boards; evidence state Measured; 72.7% support; score interval 61.7–70.1. | 131K 33K max output Exact context: 131,072 tokens; exact maximum output: 32,768 tokens | 1,395 Arena Highest supported configuration: Default; source rank 142 | $4.00 | |
46Current-model rank | 60.5 2 public benchmark observations across 2 boards; evidence state Measured; 81.2% support; score interval 60.5–67.5. | 262K 66K max output Exact context: 262,144 tokens; exact maximum output: 65,536 tokens | 1,388 Arena Highest supported configuration: Default; source rank 154 | $1.00 | |
47Current-model rank | OpenAI | 59.1 3 public benchmark observations across 3 boards; evidence state Measured; 84.2% support; score interval 59.1–65.7. | 200K 100K max output Exact context: 200,000 tokens; exact maximum output: 100,000 tokens | 1,390 Arena Highest supported configuration: Default; source rank 149 | $4.40 |
48Current-model rank | 58.6 3 public benchmark observations across 3 boards; evidence state Measured; 84.3% support; score interval 58.6–65.2. | 262K 32K max output Exact context: 262,144 tokens; exact maximum output: 32,000 tokens | 1,383 Arena Highest supported configuration: Default; source rank 159 | $0.193 | |
49Current-model rank | 58.2 2 public benchmark observations across 2 boards; evidence state Measured; 83.4% support; score interval 58.2–64.8. | 131K 8K max output Exact context: 131,072 tokens; exact maximum output: 8,192 tokens | 1,403 Arena Highest supported configuration: No thinking; source rank 132 | $1.82 | |
50Current-model rank | Z.aiOpen | 56.3 2 public benchmark observations across 2 boards; evidence state Measured; 83.3% support; score interval 56.3–62.9. | 131K 98K max output Exact context: 131,072 tokens; exact maximum output: 98,304 tokens | 1,373 Arena Highest supported configuration: Default; source rank 169 | $0.85 |