Best AI model for general assistant (2026-10)

Most capable

Gemini 4 Argon

Google

$4.00 / 1M blended

100% of leader strength

google/gemini-4-argon

Best value

Gemini 3.8 Flash

Google

$1.66 / 1M blended

92% of leader strength · 59% cheaper

google/gemini-3.8-flash

Budget

DeepSeek V4 Flash 0731

DeepSeek

$0.07 / 1M blended

74% of leader strength · 98% cheaper

deepseek/deepseek-v4-flash-0731

Best for agents

Gemini 3.8 Flash

Google

$1.66 / 1M blended

92% of leader strength · 59% cheaper

google/gemini-3.8-flash

Top 25 models for general assistant

Rows marked frontier are Pareto-optimal — nothing we track is both stronger and cheaper. "Strength" rescales the board so 100% = parity with the leader.

#	Model	Provider	Score	Strength	$/1M	Context	Tools
1	Gemini 4 Argon `google/gemini-4-argon`	Google	1526	100%	$4.00	1.048576M	tools
2	Claude Opus 5.5 `anthropic/claude-opus-5.5`	Anthropic	1516	97%	$8.00	1M	tools
3	Claude Fable 5.1 `anthropic/claude-fable-5.1`	Anthropic	1513	96%	$20.00	1M	tools
4	Claude Opus 5 `anthropic/claude-opus-5`	Anthropic	1508	95%	$10.00	1M	tools
5	Gemini 3.8 Flash `google/gemini-3.8-flash`	Google	1498	92%	$1.66	1.048576M	tools
6	Claude Fable 5 `anthropic/claude-fable-5`	Anthropic	1491	90%	$20.00	1M	tools
7	Gemini 3.7 Flash `google/gemini-3.7-flash`	Google	1490	90%	$1.50	1.048576M	tools
8	Qwen3.8 Max (0902) `qwen/qwen3.8-max-0902`	Qwen (Alibaba)	1484	88%	$3.00	1M	tools
9	Gemini 3.6 Flash `google/gemini-3.6-flash`	Google	1480	87%	$1.50	1.048576M	tools
10	Kimi K3 `moonshotai/kimi-k3`	Moonshot AI	1475	86%	$2.88	1.048576M	tools
11	GLM 5.3 `z-ai/glm-5.3`	Z.ai (Zhipu)	1474	85%	$0.41	1.048576M	tools
12	Claude Sonnet 5.5 `anthropic/claude-sonnet-5.5`	Anthropic	1473	85%	$4.00	1M	tools
13	GLM 5.3 Flash `z-ai/glm-5.3-flash`	Z.ai (Zhipu)	1472	85%	$0.24	1.048576M	tools
14	GLM 5.2 `z-ai/glm-5.2`	Z.ai (Zhipu)	1470	84%	$1.77	1.048576M	tools
15	DeepSeek V4.1 Flash `deepseek/deepseek-v4.1-flash`	DeepSeek	1464	83%	$0.27	1.048576M	tools
16	GLM 5.1 `z-ai/glm-5.1`	Z.ai (Zhipu)	1463	82%	$2.15	205K	tools
17	Claude Opus 4.8 `anthropic/claude-opus-4.8`	Anthropic	1461	82%	$10.00	1M	tools
18	Kimi K2.6 `moonshotai/kimi-k2.6`	Moonshot AI	1455	80%	$1.34	262K	tools
19	Qwen3.7 Plus `qwen/qwen3.7-plus`	Qwen (Alibaba)	1455	80%	$0.56	1M	tools
20	DeepSeek V4 Pro 0813 `deepseek/deepseek-v4-pro-0813`	DeepSeek	1451	79%	$0.99	1.048576M	tools
21	DeepSeek V4 Pro 0423 `deepseek/deepseek-v4-pro`	DeepSeek	1451	79%	$0.26	1.048576M	tools
22	Grok 4.20 `x-ai/grok-4.20`	xAI	1451	79%	$1.56	2M	tools
23	Grok 4.20 Multi-Agent `x-ai/grok-4.20-multi-agent`	xAI	1451	79%	$1.56	2M	—
24	Grok 4.5 `x-ai/grok-4.5`	xAI	1448	78%	$3.00	500K	tools
25	Kimi K2.5 `moonshotai/kimi-k2.5`	Moonshot AI	1445	77%	$1.00	262K	tools

Capability: Arena leaderboard dataset (CC-BY-4.0), Arena text, published 2026-10-22. Pricing: OpenRouter, retrieved 2026-10-25. (input x 3 + output x 1) / 4.

What this measures: Human pairwise preference votes. Measures perceived answer quality on this category, not correctness on a fixed test set. For each model this is the probability it beats the board leader in a head-to-head human vote — 0.50 means a dead tie and is the best possible value.

Get this as JSON

curl -s https://agentleaderboards.com/api/v1/route/general.json | jq .picks

No key, open CORS, refreshed daily. Set your own budget and constraints →

Other tasks

Writing & editing code · Autonomous agent work · Tool-call reliability · Following operator instructions · Recovering from shell errors · Hard reasoning · Mathematics · Precise instruction following · Long multi-turn chat · Long inputs · Creative writing · Software & IT services · Legal & government · Medicine & healthcare · Science · Business & finance · Writing & language