Task ranking
Best LLM for AI Agents
Claude Fable 5.2 ranks first for AI Agents as of November 15, 2026.[1][2][3][4] It ranks #1 on LMArena's agentic arena (score 30.4) and achieves the top tool-hallucination signal score of 1, indicating minimal fabricated tool calls in multi-step agent workflows. Claude Opus 5.5 is the next-ranked alternative. It scores 65.76% ± 3.11 (95% interval) on Terminal-Bench 4.1 (#1 of 46) at max reasoning effort, using Claude Code, measuring a model-and-agent configuration on reproducible terminal tasks.
About this recommendation
- Updated
- Evidence through
- Sources
- 15
- Revision
- v141
Decision audit
Why this result
Inspect the inputs and the computed order behind the recommendation.
Models screened
24
live candidates
Evaluation feeds
9
task-weighted
Winner coverage
41%
intended feed weight
Largest provider share
3 of 8
Anthropic
Anthropic: Claude Fable 5.2 leads the measured task comparison with 2 independent evaluation families. Community volume, release age and usage popularity add no ranking points.
Real-world track record is still developing. The recommendation can lead on independent measurements while practitioner reports accumulate.
Sources evaluated
These are the weights used in scoring. Missing measurements are omitted from each model’s average. A feed with no result for any model released in the last 60 days is stale and does not score until it catches up.
| Evaluation feed | Weight | Winner result | Field measured |
|---|---|---|---|
| Berkeley Function Callingstale | 0% | no current result | 0/24 |
| Terminal-Bench (4.1)Observed 2026-11-15 · evaluated 2026-11-13 · Claude Code; max effort · 95% interval ±2.85 | 32% | #1 | 20/24 |
| SWE-rebenchstale | 0% | no current result | 0/24 |
| LMArena AgentObserved 2026-11-11 · evaluation date not supplied · max effort | 21% | #1 | 24/24 |
| LiveBench Agentic Codingfailing; last successful check 2026-10-18 | 18% | not measured | 14/24 |
| OSWorld Verifiedstale | 0% | no current result | 0/24 |
| tau2-benchstale | 0% | no current result | 0/24 |
| LMArena Agent Task OutcomeObserved 2026-11-11 · evaluation date not supplied · max effort | 14% | #1 | 24/24 |
| LMArena Agent Tool HallucinationObserved 2026-11-11 · evaluation date not supplied · max effort | 14% | #1 | 24/24 |
Provider concentration
Each exact model is scored separately; provider identity is not a ranking input.
- Anthropic3 models
- OpenAI3 models
- Google2 models
Decision table
Every published model is shown in computed order. Practitioner sources are distinct community threads, not the citations repeated in the prose below.
Anthropic · coverage 41%
#1 LMArena Agent · #1 LMArena Agent Task Outcome
Anthropic · coverage 50%
#1 Terminal-Bench · #2 LiveBench Agentic Coding
Google · coverage 41%
#3 LMArena Agent · #3 Terminal-Bench
Anthropic · coverage 50%
#3 LMArena Agent Task Outcome · #4 LMArena Agent
OpenAI · coverage 50%
#1 LMArena Agent Tool Hallucination · #6 LMArena Agent
OpenAI · coverage 50%
#1 LMArena Agent Tool Hallucination · #5 Terminal-Bench
Google · coverage 41%
#7 LMArena Agent Task Outcome · #9 LMArena Agent
OpenAI · coverage 41%
#1 LMArena Agent Tool Hallucination · #8 LMArena Agent
Task score averages the available, task-weighted measurements. Benchmark positions compare models matched to our catalog. Models measured by at least two independent evaluator families are considered first. Missing tests, practitioner report counts, popularity and release age add no points. Coverage describes evidence breadth, not a probability of correctness. Scores within two points are shown as close under this policy; this is not a statistical confidence interval. When the leading scores are that close, the model measured by more evaluator families ranks first.
- 01
Claude Fable 5.2Anthropic · $8 in / $40 out per M tokens · 1M context
Claude Fable 5.2 ranks #1 on LMArena's agentic arena (score 30.4) and achieves the top tool-hallucination signal score of 1, indicating minimal fabricated tool calls in multi-step agent workflows.
Best when: Use for high-stakes agent deployments where task completion reliability matters: ranks #1 of 52 on explicit task-outcome signal (score 22) and #1 on tool-hallucination signal (score 1), meaning fewer failed or hallucinated tool calls.More tips
- Deploy for terminal-based agent tasks: scores 65.76% on Terminal-Bench 4.1 (#1 of 46) using Claude Code configuration.
No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
- 02
Claude Opus 5.5Anthropic · $4 in / $20 out per M tokens · 1M context
Claude Opus 5.5 scores 65.76% ± 3.11 (95% interval) on Terminal-Bench 4.1 (#1 of 46) at max reasoning effort, using Claude Code, measuring a model-and-agent configuration on reproducible terminal tasks.
Best when: Its strongest result for this task is on Terminal-Bench, which measures completing difficult repository and terminal tasks through an agent harness.No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
- 03
Gemini 4 CarbonGoogle · $2 in / $10 out per M tokens · 1.0M context
Gemini 4 Carbon ranks #3 on LMArena's agentic arena (score 17) and achieves #4 on explicit task-outcome signal (score 18.4), with strong terminal task performance at 64.55% on Terminal-Bench 4.1.
Best when: Use for terminal-based agent workflows: scores 64.55% on Terminal-Bench 4.1 (#3 of 46) at high reasoning effort using Antigravity CLI.Watch out for
- Expect higher tool-hallucination rates: ranks #25 of 52 on tool-hallucination signal (score 0.5), indicating more frequent fabricated tool calls than top performers.
- 04
Claude Sonnet 5.5Anthropic · $2 in / $10 out per M tokens · 1M context
Claude Sonnet 5.5 ranks #3 on LMArena's explicit task-outcome signal (score 19.5) and #4 on agentic arena (score 16), with a practitioner selecting it as the sole model for all agent roles in a production system.
Best when: Use as a unified agent model across diverse roles: selected as the only model for observer, tutor-editor, crop, and generator agent roles in the StudentAssistant project.More tips
- Deploy for terminal tasks with Claude Code: scores 63.94% on Terminal-Bench 4.1 (#4 of 46) at max reasoning effort.
- 05
GPT-6 AstraOpenAI · $10 in / $50 out per M tokens · 1.1M context
GPT-6 Astra ranks #1 of 52 on LMArena Agent's tool-hallucination signal (score 1), where a higher score means fewer fabricated tool calls.
Best when: Its strongest result for this task is on LMArena Agent tool hallucination, which measures calling only tools that exist, with real arguments.No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
- 06
GPT-6.1 SolOpenAI · $2 in / $10 out per M tokens · 1.1M context
GPT-6.1 Sol ranks #1 of 52 on LMArena Agent's tool-hallucination signal (score 1), where a higher score means fewer fabricated tool calls.
Best when: Its strongest result for this task is on LMArena Agent tool hallucination, which measures calling only tools that exist, with real arguments.No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.
- 07
Gemini 4 ArgonGoogle · $2 in / $10 out per M tokens · 1.0M context
Gemini 4 Argon ranks #9 on LMArena's agentic arena (score 10.7) and #7 on explicit task-outcome signal (score 13.2), with moderate terminal task performance at 55.45% on Terminal-Bench 4.1.
Best when: Use for terminal-based agent tasks: scores 55.45% on Terminal-Bench 4.1 (#10 of 46) at high reasoning effort using Antigravity CLI.Watch out for
- Watch for tool-hallucination issues: ranks #25 of 52 on tool-hallucination signal (score 0.5), indicating more frequent fabricated tool calls.
- 08
GPT-6.1 TerraOpenAI · $1 in / $5 out per M tokens · 1.1M context
GPT-6.1 Terra ranks #1 on LMArena's tool-hallucination signal (score 1) and #8 on agentic arena (score 12), though terminal task performance lags at 53.33% on Terminal-Bench 4.1.
Best when: Use for minimal tool hallucination: ranks #1 of 52 on tool-hallucination signal (score 1).Watch out for
- Expect lower task completion rates: ranks #10 of 52 on explicit task-outcome signal (score 9.7), below most competitors in this pool.
Frequently asked
- What is the best model for AI Agents?
- Claude Fable 5.2 ranks first as of November 15, 2026. Claude Fable 5.2 ranks #1 on LMArena's agentic arena (score 30.4) and achieves the top tool-hallucination signal score of 1, indicating minimal fabricated tool calls in multi-step agent workflows.[1][2][3]
- What is the best alternative to Claude Fable 5.2?
- Claude Opus 5.5 ranks second. Claude Opus 5.5 scores 65.76% ± 3.11 (95% interval) on Terminal-Bench 4.1 (#1 of 46) at max reasoning effort, using Claude Code, measuring a model-and-agent configuration on reproducible terminal tasks.[4]
Sources
- 1
“Within the 52 catalog models matched to this evaluation: Ranks #1 of 52 on LMArena Agent's tool-hallucination signal (score 1), where a higher score means fewer fabricated tool calls.”
LMArena Agent tool hallucination · Benchmark · Nov 11, 2026 - 2
“Within the 52 catalog models matched to this evaluation: Ranks #1 of 52 on LMArena's agentic arena (score 30.4), measuring tool-use and multi-step task performance from human preference.”
LMArena agentic arena · Benchmark · Nov 11, 2026 - 3
“Within the 52 catalog models matched to this evaluation: Ranks #1 of 52 on LMArena Agent's explicit task-outcome signal (score 22), measuring tasks users marked as completed.”
LMArena Agent task outcome · Benchmark · Nov 11, 2026 - 4
“Within the 46 catalog models matched to this evaluation: Scores 65.76% ± 3.11 (95% interval) on Terminal-Bench 4.1 (#1 of 46) at max reasoning effort, using Claude Code, measuring a model-and-agent configuration on reproducible terminal tasks.”
Terminal-Bench · Benchmark · Nov 15, 2026 - 5
“Within the 46 catalog models matched to this evaluation: Scores 65.76% ± 2.85 (95% interval) on Terminal-Bench 4.1 (#1 of 46) at max reasoning effort, using Claude Code, measuring a model-and-agent configuration on reproducible terminal tasks.”
Terminal-Bench · Benchmark · Nov 15, 2026 - 6
“Within the 46 catalog models matched to this evaluation: Scores 64.55% ± 3.23 (95% interval) on Terminal-Bench 4.1 (#3 of 46) at high reasoning effort, using Antigravity CLI, measuring a model-and-agent configuration on reproducible terminal tasks.”
Terminal-Bench · Benchmark · Nov 15, 2026 - 7
“Within the 52 catalog models matched to this evaluation: Ranks #25 of 52 on LMArena Agent's tool-hallucination signal (score 0.5), where a higher score means fewer fabricated tool calls.”
LMArena Agent tool hallucination · Benchmark · Nov 11, 2026 - 8
“Human decision (2026-09-29): the app uses ONLY Claude Sonnet 5.5 (claude-sonnet-5-5) for every role (observer, tutor-editor, crop, generators), replacing Opus 5.5 and Sonnet 5. Update AGENTS.md, docs/VISION.md and docs/PLAN.md accordingly (role names unchanged). Do not touch dated history in docs/runbooks/operations.md.”
MatillaM · GitHub · Sep 29, 2026 - 9
“Within the 46 catalog models matched to this evaluation: Scores 63.94% ± 3 (95% interval) on Terminal-Bench 4.1 (#4 of 46) at max reasoning effort, using Claude Code, measuring a model-and-agent configuration on reproducible terminal tasks.”
Terminal-Bench · Benchmark · Nov 15, 2026 - 10
“Within the 52 catalog models matched to this evaluation: Ranks #1 of 52 on LMArena Agent's tool-hallucination signal (score 1), where a higher score means fewer fabricated tool calls.”
LMArena Agent tool hallucination · Benchmark · Nov 11, 2026 - 11
“Within the 52 catalog models matched to this evaluation: Ranks #1 of 52 on LMArena Agent's tool-hallucination signal (score 1), where a higher score means fewer fabricated tool calls.”
LMArena Agent tool hallucination · Benchmark · Nov 11, 2026 - 12
“Within the 46 catalog models matched to this evaluation: Scores 55.45% ± 2.85 (95% interval) on Terminal-Bench 4.1 (#10 of 46) at high reasoning effort, using Antigravity CLI, measuring a model-and-agent configuration on reproducible terminal tasks.”
Terminal-Bench · Benchmark · Nov 15, 2026 - 13
“Within the 52 catalog models matched to this evaluation: Ranks #25 of 52 on LMArena Agent's tool-hallucination signal (score 0.5), where a higher score means fewer fabricated tool calls.”
LMArena Agent tool hallucination · Benchmark · Nov 11, 2026 - 14
“Within the 52 catalog models matched to this evaluation: Ranks #1 of 52 on LMArena Agent's tool-hallucination signal (score 1), where a higher score means fewer fabricated tool calls.”
LMArena Agent tool hallucination · Benchmark · Nov 11, 2026 - 15
“Within the 52 catalog models matched to this evaluation: Ranks #10 of 52 on LMArena Agent's explicit task-outcome signal (score 9.7), measuring tasks users marked as completed.”
LMArena Agent task outcome · Benchmark · Nov 11, 2026
Rankings synthesized from community evidence and open benchmarks. See methodology. Not driven by vendor marketing.