Best AI model overall right now
Claude Fable 5.2 leads the November 2026 index at 91.4, with the highest agentic (90) and knowledge (97) scores of any model here. Gemini 4 Carbon and Claude Opus 5.5 sit 3.5 points behind — and Claude Opus 5.5 does it at $20 per million output tokens against $10 for Gemini 4 Carbon. At the top, the difference between first and fifth is smaller than the difference between workloads.
Best AI model for coding in 2026
On the coding category, Claude Fable 5.2 scores highest among current models (86), ahead of Claude Opus 5.5 (82). Coding benchmarks reward getting a patch right; real engineering also needs the model to plan, run tools and recover from failures, which is what the agentic score measures. Pick the Coding preset to rank on both at once.
Best AI for agents and computer use
Claude Fable 5.2 has the strongest agentic score (90). Among the cheaper options, Gemini 4 Carbon is not yet scored on agentic tasks — worth testing for high-volume agents where per-step cost compounds.
Claude vs GPT vs Gemini: how the big three compare
| Best current model | Overall | Coding | Reasoning | $ in / out |
|---|---|---|---|---|
| Claude Fable 5.2 · Anthropic | 91.4 | 86 | 90 | $8 / $40 |
| GPT-6 Astra · OpenAI | 85.0 | 74 | 80 | $10 / $50 |
| Gemini 4 Carbon · Google | 87.9 | 81 | 83 | $2 / $10 |
Anthropic currently holds the overall lead, and Anthropic has the top coding score. Among these three, Claude Fable 5.2 is strongest on reasoning and Gemini 4 Carbon costs the least per token.
Best open-source (open-weight) LLMs
Open-weight models can be downloaded and self-hosted, which matters for data residency and per-token cost at scale. The strongest here:
- Kimi K3.1 Pro · Moonshot AI74.6
- Kimi K3 · Moonshot AI69.2
- DeepSeek V4.1 Flash · DeepSeek65.1
- Qwen3.8 Max · Alibaba64.7
- DeepSeek V4 Pro · DeepSeek62.4
- GLM-5.3 · Z.AI61.2
Cheapest frontier AI models and best value
A model is on the value frontier when nothing else is both better and cheaper. 4 models qualify: Claude Fable 5.2 ($16), Gemini 4 Carbon ($4), GPT-6.1 Terra ($2), Gemini 3.8 Flash ($1.50). Every other model is beaten on both axes by at least one of these. Switch the explorer to Price vs performance to see the line, or Cost calculator to price your own volume.
AI coding tools built on these models
These are products, not models, so they are listed rather than ranked against the table.
CursorAI code editor · Anysphere
A VS Code-based editor built around multi-file edits and fast in-IDE iteration.
Runs on: Composer 2.5 (Cursor’s own model, built on Kimi K2.5) plus Claude, GPT and Gemini models.
Composer 2.5 (May 2026): 69.3% Terminal-Bench 2.0, 79.8% SWE-Bench Multilingual — Cursor-reported.
Claude CodeTerminal coding agent · Anthropic
An agent that works in your terminal and repository: reads code, runs commands, edits files and opens pull requests.
Runs on: Claude models, including Claude Fable 5.2, Opus 5.5 and Sonnet 5.5.
OpenAI CodexCloud and CLI coding agent · OpenAI
OpenAI’s software-engineering agent, available in ChatGPT, as a CLI and in IDE extensions.
Runs on: OpenAI GPT models.
New this month
Claude Fable 5.2 from Anthropic enters at #1 with 91.4 overall. Released November 2026. New number one, with the widest uncertainty range in the top five.
Gemini 4 Carbon from Google enters at #2 with 87.9 overall. Released November 2026. Keeps 96% of the top score at a quarter of the output price.
AI benchmark categories, explained
The overall score blends five kinds of tests. What each one measures, and which model leads it today:
- Agentic
- Multi-step tool use, computer use and long-running autonomous tasks.
- Leader: Claude Fable 5.2 (90)
- Coding
- Real-repository software engineering and code generation benchmarks.
- Leader: Claude Fable 5.2 (86)
Every model at a glance
Benchmarks, price and context window for each model on the leaderboard, in rank order.
Claude Fable 5.2 benchmarks and price#1
Claude Fable 5.2 by Anthropic scores 91.4 overall, strongest on knowledge (97). API pricing is $8 per million input tokens and $40 per million output tokens, with a 1M-token context window. Its position is estimated from fewer benchmark results. Released November 2026. New number one, with the widest uncertainty range in the top five.
Gemini 4 Carbon benchmarks and price#2
Gemini 4 Carbon by Google scores 87.9 overall, strongest on reasoning (83). API pricing is $2 per million input tokens and $10 per million output tokens, with a 1M-token context window. Released November 2026. Keeps 96% of the top score at a quarter of the output price.
Claude Opus 5.5 benchmarks and price#3
Claude Opus 5.5 by Anthropic scores 85.8 overall, strongest on agentic (89). API pricing is $4 per million input tokens and $20 per million output tokens, with a 1M-token context window.
GPT-6 Astra benchmarks and price#4
GPT-6 Astra by OpenAI scores 85.0 overall, strongest on knowledge (84). API pricing is $10 per million input tokens and $50 per million output tokens, with a 1.05M-token context window.
GPT-6.1 Sol benchmarks and price#5
GPT-6.1 Sol by OpenAI scores 84.9 overall, strongest on knowledge (84). API pricing is $2 per million input tokens and $10 per million output tokens, with a 1.05M-token context window.
Gemini 4 Argon benchmarks and price#6
Gemini 4 Argon by Google scores 81.4 overall, strongest on knowledge (79). API pricing is $2 per million input tokens and $10 per million output tokens, with a 1M-token context window. Google has shipped a successor.
Claude Sonnet 5.5 benchmarks and price#7
Claude Sonnet 5.5 by Anthropic scores 81.2 overall, strongest on coding (81). API pricing is $2 per million input tokens and $10 per million output tokens, with a 1M-token context window.
Claude Fable 5.1 benchmarks and price#8
Claude Fable 5.1 by Anthropic scores 80.8 overall, strongest on knowledge (85). API pricing is $10 per million input tokens and $50 per million output tokens, with a 1M-token context window. Anthropic has shipped a successor.
Claude Fable 5 benchmarks and price#9
Claude Fable 5 by Anthropic scores 80.4 overall, strongest on knowledge (80). API pricing is $10 per million input tokens and $50 per million output tokens, with a 1M+-token context window. Anthropic has shipped a successor.
GPT-6.1 Terra benchmarks and price#10
GPT-6.1 Terra by OpenAI scores 80.1 overall, strongest on multimodal (77). API pricing is $1 per million input tokens and $5 per million output tokens, with a 1.05M-token context window. Its position is estimated from fewer benchmark results. The lowest-priced model in the frontier top 10.
Claude Opus 5 benchmarks and price#11
Claude Opus 5 by Anthropic scores 79.0 overall, strongest on knowledge (80). API pricing is $5 per million input tokens and $25 per million output tokens. Anthropic has shipped a successor.
GPT-6 Sol benchmarks and price#12
GPT-6 Sol by OpenAI scores 77.7 overall, strongest on knowledge (79). API pricing is $2 per million input tokens and $10 per million output tokens, with a 1.05M-token context window. OpenAI has shipped a successor.
Kimi K3.1 Pro benchmarks and price#13
Kimi K3.1 Pro by Moonshot AI scores 74.6 overall, strongest on knowledge (75). API pricing is $2 per million input tokens and $10 per million output tokens, with a 1.05M-token context window. Weights are open, so it can be self-hosted. Its position is estimated from fewer benchmark results. Highest-ranked open-weight model on the index. Released October 23, 2026.
GPT-5.6 Sol benchmarks and price#14
GPT-5.6 Sol by OpenAI scores 74.4 overall, strongest on knowledge (76). API pricing is $4 per million input tokens and $20 per million output tokens, with a 1.05M-token context window. OpenAI has shipped a successor.
GPT-5.5 Pro benchmarks and price#15
GPT-5.5 Pro by OpenAI scores 72.4 overall, strongest on reasoning (75). API pricing is $30 per million input tokens and $180 per million output tokens, with a 1M-token context window. Its position is estimated from fewer benchmark results.
GPT-5.6 Terra benchmarks and price#16
GPT-5.6 Terra by OpenAI scores 69.5 overall, strongest on reasoning (69). API pricing is $2 per million input tokens and $12 per million output tokens, with a 1.05M-token context window. OpenAI has shipped a successor.
Claude Opus 4.8 benchmarks and price#17
Claude Opus 4.8 by Anthropic scores 69.2 overall, strongest on knowledge (69). API pricing is $5 per million input tokens and $25 per million output tokens, with a 1M-token context window. Anthropic has shipped a successor.
Kimi K3 benchmarks and price#18
Kimi K3 by Moonshot AI scores 69.2 overall, strongest on agentic (67). API pricing is $3 per million input tokens and $15 per million output tokens, with a 1.05M-token context window. Weights are open, so it can be self-hosted. Moonshot AI has shipped a successor. Succeeded by Kimi K3.1 Pro.
Muse Wonder Full benchmarks and price#19
Muse Wonder Full by Meta scores 69.1 overall, strongest on reasoning (76). API pricing is not listed on the index, with a 1M-token context window. Its position is estimated from fewer benchmark results.
GPT-5.5 benchmarks and price#20
GPT-5.5 by OpenAI scores 68.4 overall, strongest on knowledge (70). API pricing is $5 per million input tokens and $30 per million output tokens, with a 1M-token context window. OpenAI has shipped a successor.
Claude Opus 4.7 (Adaptive) benchmarks and price#21
Claude Opus 4.7 (Adaptive) by Anthropic scores 68.1 overall, strongest on knowledge (65). API pricing is $5 per million input tokens and $25 per million output tokens, with a 1M-token context window. Anthropic has shipped a successor. Its position is estimated from fewer benchmark results.
GPT-5.4 Pro benchmarks and price#22
GPT-5.4 Pro by OpenAI scores 67.9 overall, strongest on reasoning (73). API pricing is $30 per million input tokens and $180 per million output tokens, with a 1.05M-token context window. Its position is estimated from fewer benchmark results.
Gemini 3.8 Flash benchmarks and price#23
Gemini 3.8 Flash by Google scores 66.6 overall, strongest on multimodal (70). API pricing is $0.75 per million input tokens and $3.75 per million output tokens, with a 1M-token context window.
Grok 4.6 benchmarks and price#24
Grok 4.6 by xAI scores 66.5 overall, strongest on agentic (70). API pricing is $2 per million input tokens and $6 per million output tokens, with a 500K-token context window. xAI has shipped a successor.
Gemini 3.7 Flash benchmarks and price#25
Gemini 3.7 Flash by Google scores 66.2 overall, strongest on knowledge (70). API pricing is $0.75 per million input tokens and $3.75 per million output tokens, with a 1M-token context window. Google has shipped a successor. The fastest model in this table.
DeepSeek V4.1 Flash benchmarks and price#26
DeepSeek V4.1 Flash by DeepSeek scores 65.1 overall. API pricing is not listed on the index, with a 1M-token context window. Weights are open, so it can be self-hosted. Overall score from the source’s November ranking; category scores were not listed for this row.
Qwen3.8 Max benchmarks and price#27
Qwen3.8 Max by Alibaba scores 64.7 overall. API pricing is not listed on the index, with a 1M-token context window. Weights are open, so it can be self-hosted. Overall score from the source’s November ranking; category scores were not listed for this row.
DeepSeek V4 Pro benchmarks and price#28
DeepSeek V4 Pro by DeepSeek scores 62.4 overall. API pricing is not listed on the index. Weights are open, so it can be self-hosted. Its position is estimated from fewer benchmark results. Overall score from the source’s November ranking; category scores were not listed for this row.
Qwen3.7 Max benchmarks and price#29
Qwen3.7 Max by Alibaba scores 62.1 overall. API pricing is not listed on the index. Overall score from the source’s November ranking; category scores were not listed for this row.
GLM-5.3 benchmarks and price#30
GLM-5.3 by Z.AI scores 61.2 overall. API pricing is not listed on the index, with a 1M-token context window. Weights are open, so it can be self-hosted. Overall score from the source’s November ranking; category scores were not listed for this row.
How this leaderboard works
- One source for every score. All model numbers come from BenchLM as of November 11, 2026. Vendor launch charts use different harnesses, so mixing them makes a table look precise while ranking nothing.
- Gaps stay gaps. Where the index has not scored a model on a category, the cell shows “—”. Custom weights score each model only on what was measured and show the coverage.
- Evidence labels. “Estimated” means the index scored the model on fewer benchmarks, with a wider uncertainty range. “Superseded” means its lab has shipped a successor.
- Prices are list API rates per million tokens. “$ / 1M” blends input and output 3:1, the shape of typical chat traffic.
Frequently asked questions
What is the best AI model right now (November 2026)?
Claude Fable 5.2 from Anthropic leads with an overall score of 91.4, ahead of Gemini 4 Carbon (87.9) and Claude Opus 5.5 (85.8). The gap at the top is a few points, so the best pick depends on your workload — use the presets above to rank by coding, agents or research.
What is the best AI model for coding?
Claude Fable 5.2 has the highest coding score among current models (86). For autonomous, multi-step engineering work, also weigh the agentic score, where Claude Fable 5.2 leads (90).
Which is better: Claude, GPT or Gemini?
On this index the strongest current model from each lab is Claude Fable 5.2 (91.4), GPT-6 Astra (85.0), Gemini 4 Carbon (87.9). Claude Fable 5.2 leads overall; Claude Fable 5.2 is strongest on reasoning among the three; Gemini 4 Carbon is the cheapest of them at $2 / $10 per million tokens.
What is the best open-source (open-weight) LLM?
Kimi K3.1 Pro by Moonshot AI is the highest-ranked open-weight model, scoring 74.6. 6 open-weight models appear in this leaderboard; tick “Open weights only” to see them alone.
What is the cheapest frontier AI model?
Gemini 4 Carbon is the best-scoring model on the price-performance frontier at or under $5 per million blended tokens ($4). The “Price vs performance” view plots every model so you can see the trade-off directly.
What are the newest AI models on the leaderboard?
New in November 2026: Claude Fable 5.2 by Anthropic (rank 1, score 91.4, $8 / $40 per million tokens); Gemini 4 Carbon by Google (rank 2, score 87.9, $2 / $10 per million tokens).
Where do these benchmark scores come from?
All model scores come from one independent index, BenchLM, as of November 11, 2026, so every row is measured the same way. Each model’s details link to its source page. We do not mix vendor-reported numbers from different launch posts, because cross-lab methodologies differ.
How often is the AI leaderboard updated?
We re-check the index and refresh the table when frontier models launch or scores change materially. The date at the top of the page is the date of the data, not the date you loaded it.