Model Leaderboard

Check the worth-it score before you pay. We rate the most popular AI models by real community reviews — and recommend the right one for your needs.

Compare models →

OpenAI's everyday-work model: coding, documents and multi-step tasks at about a fifth of Astra's per-task cost.

“About one-fifth of Astra's per-task cost.”

$2/$10 per 1M (cached $0.10)Per-useLast updated Nov 2026
71%Worth it

GLM 5.3

UPDATEDChat

Solid, unremarkable

“Solid and dependable. Does what it says on the box.”

$1.4/$4.4 per 1MPer-useLast updated Nov 2026
68%Worth it

Claude Opus 5.5

UPDATEDChat

Best core correctness on coding, 40% cheaper than Opus 5

“Still the most reliable model for actually shipping code. It just doesn't hallucinate the way others do.”

$4/$20 per 1M tokensPer-useLast updated Nov 2026
67%Worth it

Most loved — fast, but limits bite

“The best daily-driver model, full stop. Fast, cheap enough, and the code is clean”

$2/$10 per 1M tokensPer-use1 reviewsLast updated Nov 2026
63%Worth it

Grok 4.7

UPDATEDChat

xAI's flagship reasoning-optimized multimodal model, trained on 200,000+ GPUs, with 256K context.

$30/mo (SuperGrok Premium); free tierSubscriptionLast updated Nov 2026
63%Worth it

DeepSeek V4 Pro

UPDATEDChat

Latest DeepSeek — V4.1 with a Flash variant

“Flash hits 300 tokens/sec on real tasks — feels instant”

$0.66/$1.98 per 1M (off-peak)Per-use1 reviewsLast updated Nov 2026
62%Worth it

GPT-6.1 Terra

UPDATEDChat

OpenAI's budget mid-tier model at half the price of Sol — fast on vision and computer use.

“AWESOME release for agentic production where cost matters — best API on the market”

$1/$5 per 1M tokensPer-useLast updated Nov 2026
61%Worth it

Claude Haiku 5.5

UPDATEDChat

Anthropic's cheapest model, with an effort-control dial for reasoning.

“Noticeably smarter than GPT-6 Luna — beats Sonnet 5 in some coding benchmarks”

$0.10/$0.50 per 1M (≤100k)Per-useLast updated Nov 2026
61%Worth it

DeepSeek's open 485B-parameter Flash variant of V4.1.

“Way too fast”

$0.15/$0.60 per 1MPer-useLast updated Nov 2026
60%Worth it

GPT-6 Astra

UPDATEDChat

OpenAI's strongest model — 1M context

“Most capable model I've used this year, period. The 1M context actually works for huge codebases”

$10/$50 per 1M tokensPer-useLast updated Nov 2026
57%Worth it

Google's execution model for everyday coding and agent work — Argon-class quality at $1.50/$7.50, flat to 1M context.

“Argon-class agent capability at $1.50/$7.50 per 1M — 25% below Sonnet 5.5.”

$1.50/$7.50 per 1M (flat to 1M)Per-useLast updated Nov 2026
57%Worth it

Alibaba's fast, cheap open-weight multimodal model (Qwen 3.8 family) with 1M context, deep thinking, and agentic coding — the budget Flash sibling to Qwen 3.8 Max.

“Sparse and fast — only 6B active parameters, decoding at 150 tps on a single consumer GPU.”

$0.11 / $0.38 per 1M tokensPer-useLast updated Nov 2026
57%Worth it

Open 120B vision-language model — 5x faster for always-on agents

$0.15/$0.50 per 1MPer-useLast updated Nov 2026
52%Worth it

Doubao 2.1 Pro

UPDATEDChat

ByteDance's assistant

“ByteDance's assistant, and it's cheap.”

¥3/¥15 per 1MPer-useLast updated Nov 2026
52%Worth it

Z.ai's fast, ultra-cheap multimodal model for coding and long-horizon agent tasks (1M context).

“30M tokens processed for 50 cents — flagship-level output at a fraction of the price.”

$0.03/$0.08 per 1M tokensPer-useLast updated Nov 2026
51%Worth it

Tencent's open-source MoE flagship — 770B total/49B active params, 1M context, top open-source tier for math and agentic work.

“Horizon Math score is the best among all open models, second only to GPT-5.6 Sol.”

¥6/¥18 per 1M ($0.83/$2.50)Per-use1 reviewsLast updated Nov 2026
49%Not worth it

Muse Wonder Full

UPDATEDChat

#3 on the Intelligence Index (56)

“Where did this come from? It appeared out of nowhere and it's sitting at #3 on the index”

$50/moSubscriptionLast updated Nov 2026
47%Not worth it

Mistral Large 4

UPDATEDChat

European open-weights flagship with aggressive preview pricing.

“High-effort mode renders the pelican correctly — feet connect to pedals”

$0.70/$2.10 per 1MPer-useLast updated Nov 2026
47%Not worth it

Qwen4-Max

NEWChat

Alibaba's 2.4T-parameter MoE flagship (71B active), native vision, 1M-token context — 6th on Code Arena.

“2.4T total parameters with only 71B active per token — lower inference cost at the same scale.”

$0.075/M input (cache hit)Per-useLast updated Nov 2026
47%Not worth it

Qwen 3.8 Max

UPDATEDChat

Alibaba's flagship 2.4T-parameter MoE model (95B active) with a 1M-token context, native image/video understanding, and strong agentic coding; first Max-class Qwen with open weights.

“Alibaba's best model yet, and the Chinese + English mix is finally balanced.”

$2/$6 per 1M tokensPer-useLast updated Nov 2026
43%Not worth it

OpenAI's professional-work model: financial analysis, legal research, slides and complex coding — 20% cheaper than GPT-6 Astra.

“Agents Last Exam +5.2 over GPT-6 Astra and +5.5 over Claude Fable 5.2 (OpenAI-reported).”

$8/$40 per 1M (cached $0.40)Per-useLast updated Nov 2026
42%Not worth it

GPT 6.1 Sol

UPDATEDChat

Bills run high on agent tasks

“The actual value king from OpenAI. Does 90% of what Astra does for way less.”

$2/$10 per 1M tokensPer-use1 reviewsLast updated Nov 2026
41%Not worth it

Gemini 4 Argon

UPDATEDChat

Cheapest frontier model, true multimodal

“$2/$10 per million tokens is a steal for what you get. Cheapest frontier model by a mile.”

$2/$10 per 1M (→$4/$20 Dec 1)Per-useLast updated Nov 2026
40%Not worth it

Google's strongest frontier model — tops Intelligence Index v5, state-of-the-art cybersecurity + long coding. Gated to cyber defenders.

“More capable and the same price — I can at least appreciate that.”

$2/$10 per 1M (→$4/$20 Dec 1)Per-useLast updated Nov 2026
39%Not worth it

MiniMax's fast, cheap text/language model (M3.1 Flash) for coding, reasoning, and agentic tasks.

“Cuts agent task cost by 70%, which is the whole point.”

$0.15/$0.60 per 1MPer-useLast updated Nov 2026
37%Not worth it

Reflection Beam

UPDATEDChat

New US open-weights model

“New US open-weights model from Reflection — their first launch.”

$0.40/$1.20 per 1MPer-useLast updated Nov 2026
36%Not worth it

Kimi K3.1 Pro

UPDATEDChat

2.8T params, beats GPT-6 Astra on Terminal-Bench

“The 1M context actually works. I dumped a whole codebase in and it kept track of everything”

$2/$10 per 1MPer-use1 reviewsLast updated Nov 2026
31%Not worth it

Claude Max

UPDATEDChat

Expensive, refund complaints

“The model quality is there, but the Max pricing is brutal”

$100/moSubscription1 reviewsLast updated Nov 2026
31%Not worth it

OpenAI's cybersecurity reasoning model — #1 on ExploitBench, helped find 81 Firefox flaws and 948 Linux exploits. Gated to approved orgs (Daybreak Red).

“An AI that can discover real security vulnerabilities in the Linux kernel — this is the kind of progress that really matters.”

$20/$100 per 1M tokensPer-useLast updated Nov 2026
27%Not worth it