Skip to main content

Human ratings

Legal and government (human ratings)

Among the occupational fields questions are sorted into, a category computed only from questions in legal and government.

How to read this table

  • Higher rows are rated higher. A range like “1–2” means the gap is within the margin of error.
  • The rating comes from people comparing two answers and picking the better one.
  • At a gap of 50, the higher model is picked about 57% of the time; 100 means 64%, 200 means 76%.
  • Prices are a guide per million tokens, reading then writing.
  • Length is how much text the model can read at once (tokens).

± range of the top model

RankModelDeveloperratingrating scale 1400 to 1550Week
change
OverallInput $
/1M
Output $
/1M
ContextSupportsWeightsReleasedPrice and specs
1–24Muse Spark 1.2MetaRating 1523 ±21−131491rank 6–25$1.25$4.251.05MMeta Released 2026-⁠08 Week change −13 Price $1.25 / $4.25 Length 1.05M
1–36Gemini 4 ArgonGoogleRating 1514 ±22−211517rank 1–3$2$101.05MGoogle Released 2026-⁠09 Week change −21 Price $2 / $10 Length 1.05M
1–49Muse WonderMetaRating 1511 ±27New1509rank 1–8$4$201.05MMeta Released 2026-⁠10 Price $4 / $20 Length 1.05M
1–25Claude Opus 4.7AnthropicRating 1511 ±9−11501rank 2–10$5$251MAnthropic Released 2026-⁠04 Week change −1 Price $5 / $25 Length 1M
1–28Claude Fable 5AnthropicRating 1510 ±10−21503rank 2–9$10$501MAnthropic Released 2026-⁠06 Week change −2 Price $10 / $50 Length 1M
1–25Claude Opus 4.6AnthropicRating 1509 ±8−11503rank 2–9$5$251MAnthropic Released 2026-⁠02 Week change −1 Price $5 / $25 Length 1M
1–30Kimi K3⁠MoonshotRating 1509 ±12−11487rank 8–27$3$151.05MMoonshot Released 2026-⁠07 Week change −1 Price $3 / $15 Length 1.05M
1–37Claude Opus 5AnthropicRating 1504 ±12+11491rank 8–22$5$251MAnthropic Released 2026-⁠07 Week change +1 Price $5 / $25 Length 1M
1–51Claude Opus 5.5AnthropicRating 1503 ±21+81509rank 1–7$4$201MAnthropic Released 2026-⁠09 Week change +8 Price $4 / $20 Length 1M
1–43Muse Spark 1.3MetaRating 1502 ±15−21496rank 3–15$1.25$4.251.05MMeta Released 2026-⁠09 Week change −2 Price $1.25 / $4.25 Length 1.05M
1–39Gemini 3 ProGoogleRating 1501 ±1101486rank 10–29——1.05MGoogle Released 2025-⁠11 Price — Length 1.05M
1–54Muse SparkMetaRating 1500 ±19+11489rank 8–27———Meta Released 2026-⁠04 Week change +1 Price — Length —
1–37Gemini 3.1 Pro (preview)GoogleRating 1498 ±701487rank 10–26$2$121.05MGoogle Released 2026-⁠02 Price $2 / $12 Length 1.05M
1–41Claude Opus 4.8AnthropicRating 1498 ±901481rank 12–36$5$251MAnthropic Released 2026-⁠05 Price $5 / $25 Length 1M
1–43Muse Spark 1.1MetaRating 1497 ±10+21490rank 8–23$1.25$4.251MMeta Released 2026-⁠07 Week change +2 Price $1.25 / $4.25 Length 1M
1–44GPT-5.6 SolOpenAIRating 1497 ±11−11485rank 10–30$4$201.05MOpenAI Released 2026-⁠07 Week change −1 Price $4 / $20 Length 1.05M
1–70GPT-6.1 SolOpenAIRating 1497 ±23−11480rank 10–41$2$101.05MOpenAI Released 2026-⁠09 Week change −1 Price $2 / $10 Length 1.05M
1–47Gemini 3.7 FlashGoogleRating 1496 ±12+41490rank 8–25$0.75$3.751.05MGoogle Released 2026-⁠08 Week change +4 Price $0.75 / $3.75 Length 1.05M
1–44GPT-5.5OpenAIRating 1495 ±901482rank 12–36$5$301.05MOpenAI Released 2026-⁠04 Price $5 / $30 Length 1.05M
4–53GLM-5.2⁠Z.aiRating 1491 ±1001475rank 19–43$1.4$4.41MZ.ai Released 2026-⁠06 Price $1.4 / $4.4 Length 1M
4–55Gemini 3.8 FlashGoogleRating 1491 ±10−11496rank 4–15$0.75$3.751.05MGoogle Released 2026-⁠09 Week change −1 Price $0.75 / $3.75 Length 1.05M
1–70Claude Fable 5.1AnthropicRating 1491 ±17−11503rank 2–10$10$501MAnthropic Released 2026-⁠09 Week change −1 Price $10 / $50 Length 1M
5–53GPT-5.4OpenAIRating 1490 ±901475rank 19–42$2.5$151.05MOpenAI Released 2026-⁠03 Price $2.5 / $15 Length 1.05M
1–75GPT-6 AstraOpenAIRating 1489 ±17+31476rank 17–44$10$501.05MOpenAI Released 2026-⁠09 Week change +3 Price $10 / $50 Length 1.05M
1–63Qwen3.8 MaxAlibabaRating 1489 ±13−11483rank 10–36$2$61MAlibaba Released 2026-⁠08 Week change −1 Price $2 / $6 Length 1M
5–64Claude Opus 4.5AnthropicRating 1487 ±1101474rank 23–46$5$25200KAnthropic Released 2025-⁠11 Price $5 / $25 Length 200K
6–63Claude Sonnet 4.6AnthropicRating 1484 ±901472rank 24–49$3$151MAnthropic Released 2026-⁠02 Price $3 / $15 Length 1M
6–69Gemini 3.5 FlashGoogleRating 1484 ±1001477rank 18–41$1.5$91.05MGoogle Released 2026-⁠05 Price $1.5 / $9 Length 1.05M
6–75Gemini 3 FlashGoogleRating 1483 ±1201473rank 23–49——1.05MGoogle Released 2025-⁠12 Price — Length 1.05M
1–89Claude Sonnet 5.5AnthropicRating 1483 ±22−121478rank 14–44$2$101MAnthropic Released 2026-⁠09 Week change −12 Price $2 / $10 Length 1M
6–77GPT-5.2 ChatOpenAIRating 1482 ±12+11476rank 19–42——128KOpenAI Released 2026-⁠02 Week change +1 Price — Length 128K
7–77Gemini 3.6 FlashGoogleRating 1481 ±11−11483rank 10–35$0.75$3.751.05MGoogle Released 2026-⁠07 Week change −1 Price $0.75 / $3.75 Length 1.05M
4–90DeepSeek-V4 Pro (2026-⁠08)⁠DeepSeekRating 1480 ±20+11465rank 30–64$0.66$1.981MDeepSeek Released 2026-⁠08 Week change +1 Price $0.66 / $1.98 Length 1M
1–110Qwen3.7 Max (preview)AlibabaRating 1480 ±3401474rank 15–53————Alibaba Released — Price — Length —
13–77Grok 4.20 Beta Reasoning (0309)xAIRating 1478 ±901472rank 24–50——2M—xAI Released — Price — Length 2M
10–81Grok 4.5xAIRating 1478 ±1101466rank 33–55$2$6500KxAI Released 2026-⁠07 Price $2 / $6 Length 500K
10–88GPT-5.5 InstantOpenAIRating 1477 ±1301473rank 20–50———OpenAI Released 2026-⁠05 Price — Length —
15–86GPT-5.6 TerraOpenAIRating 1475 ±1001467rank 28–53$2$121.05MOpenAI Released 2026-⁠07 Price $2 / $12 Length 1.05M
15–84Grok 4.1xAIRating 1475 ±901465rank 38–53——200KxAI Released 2025-⁠11 Price — Length 200K
17–83MiMo-V2.5 Pro⁠XiaomiRating 1475 ±801466rank 34–53$0.435$0.871.05MXiaomi Released 2026-⁠04 Price $0.435 / $0.87 Length 1.05M
How to read this page
  • These ratings exclude the effect of response length and formatting. Differences where the ± ranges overlap are within the margin of error, so about the same.
  • In the Supports column the icons are, from left: can use tools / can read images / can answer in a fixed format (JSON); faded means not supported or unknown.
  • Open-weights models have their contents (the weights) published, so you can also run them on your own computer or server. Models without the open-weights mark are normally used through the developer's own services (apps or API); for some models the weights status is unknown.
  • Names such as ChatGPT next to a model are the consumer services its developer offers. Check on the service itself whether that model is available there.

Data: Arena, Epoch AI, Swallow LLM team (CC BY 4.0), Models.dev (MIT), updated 2026-⁠10-⁠24, data sources