1stTop pick
Researched 10d agoClaude Opus 4.7
- GPQA Diamond
- 94.2%
- Output (from)
- $25.00 / 1M
The Best Mainstream LLM APIs, Ranked (2026)
Last refreshed 2026-06-03. Next refresh: weekly.
The best mainstream APIs, ranked by capability first: GPQA Diamond, MMLU fallback, then lowest tracked input price.
Verdict
Use Claude Opus 4.7 for mainstream API work today.
GPT-5.5 is the runner-up, 0.6 points back on GPQA Diamond.
2ndShortlist
Researched 1d agoGPT-5.5
- GPQA Diamond
- 93.6%
- Output (from)
- $30.00 / 1M
3rdShortlist
Researched 7d agoClaude Opus 4.8
- GPQA Diamond
- 93.6%
- Output (from)
- $25.00 / 1M
How we rank
Mainstream API picks now lead with capability: GPQA Diamond first, MMLU as fallback, then price only as the tie-break.
- Eligibility — Chat/completion models that pass the generic API leaderboard filter (no embeddings/rerankers/modality SKUs).
- Primary ranking — GPQA Diamond, then MMLU when GPQA is missing or tied.
- Variant collapse — We keep one row per model family (`familySlug` + parameter tier). When headline scores tie within ±0.5 pt (±10 Elo on Chatbot Arena), we pick the canonical SKU by lowest tracked input price, then GA over preview or limited access, then newest `release`. A folded sibling within the benchmark noise band can show a "Tied within margin" chip on that score cell.
- Price tie-break — Lower tracked input $/1M wins only after the capability scores are exhausted.
- Pricing — Rates reflect tracked provider rows — spot/check enterprise tiers separately.
| # | Model | Input $/1M | Output $/1M | |
|---|---|---|---|---|
| 1 | Gemini 3.1 Pro Preview PreviewVisionTools Capability signal: GPQA Diamond 94.3% | $2.00 | $12.00 | |
| 2 | Claude Opus 4.7 ReasoningVisionTools Capability signal: GPQA Diamond 94.2% | $5.00 | $25.00 | |
| 3 | GPT-5.5 ReasoningVisionTools Capability signal: GPQA Diamond 93.6% | $5.00 | $30.00 | |
| 4 | Claude Opus 4.8 ReasoningVisionTools Capability signal: GPQA Diamond 93.6% | $5.00 | $25.00 | |
| 5 | GPT-5.5 Pro ReasoningVisionTools Capability signal: GPQA Diamond 93.6% | $30.00 | $180.00 | |
| 6 | Qwen3.7-Max ReasoningTools Capability signal: GPQA Diamond 92.4% | $1.25 | $3.75 | |
| 7 | GPT-5.4 ReasoningTools Capability signal: GPQA Diamond 92% | $2.50 | $15.00 | |
| 8 | Gemini 3 Pro VisionTools Capability signal: GPQA Diamond 91.9% | $1.25 | $5.00 | |
| 9 | Claude Opus 4.6 ReasoningVisionTools Capability signal: GPQA Diamond 91.3% | $5.00 | $25.00 | |
| 10 | Kimi K2.6 ReasoningVisionTools Capability signal: GPQA Diamond 90.5% | $0.73 | $3.40 | |
| 11 | Gemini 3 Flash PreviewVisionTools Capability signal: GPQA Diamond 90.4% | $0.50 | $3.00 | |
| 12 | DeepSeek V4 Pro ReasoningTools Capability signal: GPQA Diamond 90.1% | $0.43 | $0.87 | |
| 13 | Grok 4.3 ReasoningVisionTools Capability signal: GPQA Diamond 90.1% | $1.25 | $2.50 | |
| 14 | Claude Sonnet 4.6 ReasoningVisionTools Capability signal: GPQA Diamond 89.9% | $3.00 | $15.00 | |
| 15 | Qwen3.5-397B-A17B ReasoningTools Capability signal: GPQA Diamond 89.3% | $0.39 | $2.34 | |
| 16 | Trinity-Large-Thinking ReasoningTools Capability signal: GPQA Diamond 89.2% | $0.22 | $0.85 | |
| 17 | ByteDance Doubao Seed 2.0 Pro VisionTools Capability signal: GPQA Diamond 88.9% | $0.47 | $2.37 | |
| 18 | GPT-5 ReasoningVisionTools Capability signal: GPQA Diamond 88.4% | $1.25 | $10.00 | |
| 19 | DeepSeek V4 Flash ReasoningTools Capability signal: GPQA Diamond 88.1% | $0.10 | $0.20 | |
| 20 | Grok 4.20 ReasoningVisionTools Capability signal: GPQA Diamond 88% | $1.25 | $2.50 |
Honorable mentions
Next seats in this ranking. Lines below are from each model's stored description in LLMReference seed data—spot-check the model page before relying on a capability claim.
GPT-5.5 Pro is OpenAI's premium variant of GPT-5.5, released April 23, 2026. Targets large quality gains for business, legal, education, and data science use cases. Scores 39.6% on FrontierMath Tier 4 (postdoctoral-level math problems), compared to 22.9% for Claude Opus 4.7. Priced at 6× the standard GPT-5.5 API rate. Available to ChatGPT subscribers and via API.
93.6%
GPQA Diamond
Qwen3.7-Max is Alibaba's flagship agentic reasoning model, announced at the Alibaba Cloud Summit on May 20, 2026. It features a 1M-token context window, extended-thinking (chain-of-thought) mode, and is designed for long-horizon autonomous tasks including coding, debugging, and multi-step workflows. The model is text-only (no vision input) and is available via Alibaba Cloud Model Studio (DashScope). Closed-weight; no open-source weights have been released.
92.4%
GPQA Diamond
- #6GPT-5.4
GPT-5.4 is OpenAI's flagship frontier reasoning model, released March 5, 2026. It incorporates advances from GPT-5.3-Codex for coding and agentic workflows, and adds 'Thinking' mode with editable reasoning plans. Key capabilities include computer use (navigating interfaces via Playwright), image understanding and generation integration, full-stack web app generation, tool calling, and deep research. Knowledge cutoff is August 31, 2025. Model ID: gpt-5.4.
92%
GPQA Diamond
Compare Top Picks
Side-by-side comparison of the top picks by price, benchmark, and API access.
Browse Other Categories
Best LLMs for Code GenerationBest LLMs for RAGBest AI Agents & Agentic ModelsBest LLMs for ClassificationBest Open Source LLMsBest Multimodal / Vision LLMsBest LLMs for Reasoning & MathBest Small Language Models (SLMs)Best LLMs for Function Calling & Tool UseCheapest LLM APIs You Can Call Right NowBest Long Context LLMsBest LLMs for EnterpriseBest Free LLMs You Can Use Right NowBest LLMs for WritingBest LLMs for MarketingBest LLMs for Customer Support