New: Agent Index for knowledge work is live.Explore the index ↗
Compare models

Best LLMs for business deliverables

AA-Briefcase

AA-Briefcase evaluates agents producing spreadsheets, presentations, memos and other professional deliverables across 91 tasks. Its combined Elo incorporates rubric performance, analytical quality and presentation. Each task runs independently, and the public Lite example is excluded.

Updated: Data as of:
Widely usedSaturation: not applicable
Official benchmark
AA-Briefcase results. Each row identifies its benchmark version and evaluation track. Scores from different versions are not directly comparable. Best effort from the latest stored report date per model and version.
Model / creatorVersionReported score
Claude Sonnet 5.5Anthropic · Proprietary1.1 / combined Elo1,814 Elo
Claude Opus 5.5Anthropic · Proprietary1.1 / combined Elo1,801 Elo
Kimi K3.1 ProMoonshot AI · Open source / open weights1.1 / combined Elo1,675.78 Elo
Claude Fable 5.1Anthropic · Proprietary1.1 / combined Elo1,670 Elo
Muse Wonder 1.0Meta · Proprietary1.1 / combined Elo1,659 Elo
Claude Opus 5Anthropic · Proprietary1.1 / combined Elo1,656 Elo
Grok 4.7xAI · Proprietary1.1 / combined Elo1,638 Elo
Gemini 4 CarbonGoogle · Proprietary1.1 / combined Elo1,610 Elo
Muse Spark 1.3Meta · Proprietary1.1 / combined Elo1,581 Elo
Claude Haiku 5.5Anthropic · Proprietary1.1 / combined Elo1,576 Elo
DeepSeek V4.1 Flash 1017DeepSeek · Open source / open weights1.1 / combined Elo1,570.42 Elo
GPT-6 AstraOpenAI · Proprietary1.1 / combined Elo1,569 Elo
GPT-6.1 SolOpenAI · Proprietary1.1 / combined Elo1,555 Elo
MiMo-V2.6-ProXiaomi · Open source / open weights1.1 / combined Elo1,512 Elo
GLM-5.3Z.ai · Open source / open weights1.1 / combined Elo1,504.57 Elo
Kimi K3Moonshot AI · Open source / open weights1.1 / combined Elo1,498.42 Elo
Gemini 4 ArgonGoogle · Proprietary1.1 / combined Elo1,484 Elo
GPT-5.6 SolOpenAI · Proprietary1.1 / combined Elo1,479 Elo
MiniMax M3.1 FlashMiniMax · Open source / open weights1.1 / combined Elo1,474.82 Elo
GLM-5.3-FlashZ.ai · Open source / open weights1.1 / combined Elo1,450.23 Elo
Step 5 PreviewStepFun · Open source / open weights1.1 / combined Elo1,425 Elo
DeepSeek V4.1 FlashDeepSeek · Open source / open weights1.1 / combined Elo1,416.94 Elo
GPT-6.1 TerraOpenAI · Proprietary1.1 / combined Elo1,402 Elo
Qwen3.8 27BQwen · Open source / open weights1.1 / combined Elo1,393 Elo
Claude Sonnet 5Anthropic · Proprietary1.1 / combined Elo1,357 Elo
GPT-5.6 LunaOpenAI · Proprietary1.1 / combined Elo1,339 Elo
GPT-6 LunaOpenAI · Proprietary1.1 / combined Elo1,337 Elo
GPT-5.6 TerraOpenAI · Proprietary1.1 / combined Elo1,333 Elo
Muse Spark 1.2Meta · Proprietary1.1 / combined Elo1,325 Elo
Claude Opus 4.8Anthropic · Proprietary1.1 / combined Elo1,320 Elo
Muse Spark OpenMeta · Open source / open weights1.1 / combined Elo1,279 Elo
DeepSeek V4 Pro 0813DeepSeek · Open source / open weights1.1 / combined Elo1,254.44 Elo
Claude Opus 4.7Anthropic · Proprietary1.1 / combined Elo1,252 Elo
GLM-5.2Z.ai · Open source / open weights1.1 / combined Elo1,227.23 Elo
Gemini 3.8 FlashGoogle · Proprietary1.1 / combined Elo1,203 Elo
GPT-5.5OpenAI · Proprietary1.1 / combined Elo1,136 Elo
MiniMax M3MiniMax · Open source / open weights1.1 / combined Elo1,089.34 Elo
Claude Sonnet 4.6Anthropic · Proprietary1.1 / combined Elo1,062 Elo
DeepSeek V4 Flash 0420DeepSeek · Open source / open weights1.1 / combined Elo1,041.27 Elo
DeepSeek V4 FlashDeepSeek · Open source / open weights1.1 / combined Elo1,041 Elo
Nemotron 3.5 Super VLNVIDIA · Open source / open weights1.1 / combined Elo999 Elo
GLM-5.1Z.ai · Open source / open weights1.1 / combined Elo965 Elo
DeepSeek V4 Pro 0424DeepSeek · Open source / open weights1.1 / combined Elo921.93 Elo
Muse Spark 1.1Meta · Proprietary1.1 / combined Elo851 Elo
InklingThinking Machines · Open source / open weights1.1 / combined Elo833 Elo
Kimi K2.6Moonshot AI · Open source / open weights1.1 / combined Elo820.03 Elo
GPT-5.4 miniOpenAI · Proprietary1.1 / combined Elo720 Elo
MiniMax M2.7MiniMax · Open source / open weights1.1 / combined Elo719 Elo
Claude Sonnet 4.5Anthropic · Proprietary1.1 / combined Elo714 Elo
Muse SparkMeta · Proprietary1.1 / combined Elo646 Elo
Qwen3.5-397B-A17BQwen · Open source / open weights1.1 / combined Elo563 Elo
Gemini 3.1 ProGoogle · Proprietary1.1 / combined Elo457 Elo
Qwen 3.5 9BQwen · Open source / open weights1.1 / combined Elo341 Elo
DeepSeek R1 (Jan 2025)DeepSeek · Open source / open weights1.1 / combined Elo128.91 Elo
gpt-oss-120bOpenAI · Open source / open weights1.1 / combined Elo0 Elo
gpt-oss-20bOpenAI · Open source / open weights1.1 / combined Elo0 Elo
Qwen3-14BQwen · Open source / open weights1.1 / combined Elo0 Elo
View all 57 published reports
1.1 / combined Elo

selection: publisher default displayed configuration · Claude Sonnet 5.5 (max with fallback)

1.1 / combined Elo

selection: publisher default displayed configuration · Claude Opus 5.5 (max with fallback)

Kimi K3.1 Pro1,675.78 Elo
1.1 / combined Elo

selection: best reported configuration per release · Kimi K3.1 Pro · tools: benchmark-defined

1.1 / combined Elo

selection: publisher default displayed configuration · Claude Fable 5.1 (max with fallback)

1.1 / combined Elo

selection: best reported configuration per release · Muse Wonder (max) · tools: benchmark-defined

Claude Opus 51,656 Elo
1.1 / combined Elo

selection: best reported configuration per release · Claude Opus 5 (max) · tools: benchmark-defined

Grok 4.71,638 Elo
1.1 / combined Elo

selection: publisher default displayed configuration · Grok 4.7 (xhigh)

1.1 / combined Elo

selection: publisher default displayed configuration · Gemini 4 Carbon (high)

1.1 / combined Elo

selection: best reported configuration per release · Muse Spark 1.3 (max) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Claude Haiku 5.5 (max) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · DeepSeek V4.1 Flash 1017 (Max) · tools: benchmark-defined

GPT-6 Astra1,569 Elo
1.1 / combined Elo

selection: publisher default displayed configuration · GPT-6 Astra (max)

GPT-6.1 Sol1,555 Elo
1.1 / combined Elo

selection: publisher default displayed configuration · GPT-6.1 Sol (max)

MiMo-V2.6-Pro1,512 Elo
1.1 / combined Elo

selection: best reported configuration per release · MiMo-V2.6-Pro · tools: benchmark-defined

GLM-5.31,504.57 Elo
1.1 / combined Elo

selection: best reported configuration per release · GLM-5.3 (Max) · tools: benchmark-defined

Kimi K31,498.42 Elo
1.1 / combined Elo

selection: best reported configuration per release · Kimi K3 (Max) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Gemini 4 Argon (high) · tools: benchmark-defined

GPT-5.6 Sol1,479 Elo
1.1 / combined Elo

selection: best reported configuration per release · GPT-5.6 Sol (max) · tools: benchmark-defined

MiniMax M3.1 Flash1,474.82 Elo
1.1 / combined Elo

selection: best reported configuration per release · MiniMax-M3.1-Flash · tools: benchmark-defined

GLM-5.3-Flash1,450.23 Elo
1.1 / combined Elo

selection: best reported configuration per release · GLM-5.3-Flash · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Step 5 Preview · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · DeepSeek V4.1 Flash (Max) · tools: benchmark-defined

GPT-6.1 Terra1,402 Elo
1.1 / combined Elo

selection: best reported configuration per release · GPT-6.1 Terra (max) · tools: benchmark-defined

Qwen3.8 27B1,393 Elo
1.1 / combined Elo

selection: best reported configuration per release · Qwen3.8 27B (xhigh) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Claude Sonnet 5 (max) · tools: benchmark-defined

GPT-5.6 Luna1,339 Elo
1.1 / combined Elo

selection: best reported configuration per release · GPT-5.6 Luna (max) · tools: benchmark-defined

GPT-6 Luna1,337 Elo
1.1 / combined Elo

selection: best reported configuration per release · GPT-6 Luna (max) · tools: benchmark-defined

GPT-5.6 Terra1,333 Elo
1.1 / combined Elo

selection: best reported configuration per release · GPT-5.6 Terra (max) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Muse Spark 1.2 (xhigh) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Claude Opus 4.8 (max) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Muse Spark Open (xhigh) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · DeepSeek V4 Pro 0813 (Max) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Claude Opus 4.7 (max) · tools: benchmark-defined

GLM-5.21,227.23 Elo
1.1 / combined Elo

selection: best reported configuration per release · GLM-5.2 (Max) · tools: benchmark-defined

1.1 / combined Elo

selection: publisher default displayed configuration · Gemini 3.8 Flash (high)

GPT-5.51,136 Elo
1.1 / combined Elo

selection: best reported configuration per release · GPT-5.5 (xhigh) · tools: benchmark-defined

MiniMax M31,089.34 Elo
1.1 / combined Elo

selection: best reported configuration per release · MiniMax-M3 · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Claude Sonnet 4.6 (max) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · DeepSeek V4 Flash 0420 (High) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · DeepSeek V4 Flash (high) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Nemotron 3.5 Super VL · tools: benchmark-defined

GLM-5.1965 Elo
1.1 / combined Elo

selection: best reported configuration per release · GLM-5.1 · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · DeepSeek V4 Pro 0424 (Max) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Muse Spark 1.1 (xhigh) · tools: benchmark-defined

Inkling833 Elo
1.1 / combined Elo

selection: best reported configuration per release · Inkling (xhigh) · tools: benchmark-defined

Kimi K2.6820.03 Elo
1.1 / combined Elo

selection: best reported configuration per release · Kimi K2.6 (Reasoning) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · GPT-5.4 mini (xhigh) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · MiniMax-M2.7 · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Claude 4.5 Sonnet · tools: benchmark-defined

Muse Spark646 Elo
1.1 / combined Elo

selection: best reported configuration per release · Muse Spark · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Qwen3.5 397B A17B · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Gemini 3.1 Pro Preview · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Qwen3.5 9B · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · DeepSeek R1 (Jan '25) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · Qwen3 14B · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · gpt-oss-120b (High) · tools: benchmark-defined

1.1 / combined Elo

selection: best reported configuration per release · gpt-oss-20b (High) · tools: benchmark-defined