Task ranking

Best LLM for Coding

Claude Fable 5.2 ranks first for Coding as of November 15, 2026.[1][2] It 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. 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
9
Revision
v149

Decision audit

Why this result

Inspect the inputs and the computed order behind the recommendation.

Models screened

24

live candidates

Evaluation feeds

7

task-weighted

Winner coverage

47%

intended feed weight

Largest provider share

3 of 8

Anthropic

Established source breadth. 5 citation families and 1 practitioner families provide attributed capability reports; 0 cautionary threads retained. The largest citation family contributes 31%.

Anthropic: Claude Fable 5.2 leads the measured task comparison with 3 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.

winner: Claude Fable 5.2
Evaluation feedWeightWinner resultField measured
LMArena WebDevObserved 2026-11-11 · evaluation date not supplied · xhigh effort
31%
#122/24
LiveBench Codingfailing; last successful check 2026-10-18
28%
not measured12/24
SWE-rebenchstale
0%
no current result0/24
Design Arena CodingObserved 2026-11-15 · evaluation date not supplied · anthropic/claude-fable-5.2-20261109
24%
#321/24
Aider Polyglotstale
0%
no current result0/24
Terminal-Bench (4.1)Observed 2026-11-15 · evaluated 2026-11-13 · Claude Code; max effort · 95% interval ±2.85
17%
#121/24
SWE-bench Verifiedstale
0%
no current result0/24

Provider concentration

Each exact model is scored separately; provider identity is not a ranking input.

Anthropic38%
  • Anthropic3 models
  • Google3 models
  • Moonshot AI1 model
  • OpenAI1 model

Decision table

Every published model is shown in computed order. Practitioner sources are distinct community threads, not the citations repeated in the prose below.

  1. Anthropic · coverage 47%

    #1 LMArena WebDev · #1 Terminal-Bench

  2. Anthropic · coverage 64%

    #1 Terminal-Bench · #2 Design Arena Coding

  3. Anthropic · coverage 64%

    #1 LiveBench Coding · #3 LMArena WebDev

  4. Google · coverage 47%

    #3 Terminal-Bench · #4 LMArena WebDev

  5. Moonshot AI · coverage 47%

    #7 Design Arena Coding · #7 Terminal-Bench

  6. Google · coverage 47%

    #8 LMArena WebDev · #9 Terminal-Bench

  7. Google · coverage 47%

    #10 Terminal-Bench · #11 LMArena WebDev

  8. OpenAI · coverage 64%

    #4 Design Arena Coding · #5 Terminal-Bench

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.

  1. 01

    Claude Fable 5.2Anthropic · $8 in / $40 out per M tokens · 1M context

    Claude Fable 5.2 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.

    Best when: Its strongest result for this task is on Terminal-Bench, which measures completing difficult repository and terminal tasks through an agent harness.
    Source 1
    “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-BenchOpen original ↗
  2. 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.
    Source 2
    “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-BenchOpen original ↗
  3. 03

    Claude Sonnet 5.5Anthropic · $2 in / $10 out per M tokens · 1M context

    Claude Sonnet 5.5 scores 91.37% on LiveBench Coding (#1 of 59), an objective evaluation of code generation and completion.

    Best when: Its strongest result for this task is on LiveBench Coding, which measures code generation and completion.
    Source 4
    “Within the 59 catalog models matched to this evaluation: Scores 91.37% on LiveBench Coding (#1 of 59), an objective evaluation of code generation and completion.”
    LiveBench CodingOpen original ↗

    No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.

  4. 04

    Gemini 4 CarbonGoogle · $2 in / $10 out per M tokens · 1.0M context

    Ranks #8 on Design Arena coding category (Elo 1348) and #4 on LMArena WebDev coding arena (Elo 1768), with 64.55% on Terminal-Bench 4.1 using Antigravity CLI.

    Best when: Use for web development where it ranks fourth among 102 models on human preference votes.
    Source 5
    “Within the 102 catalog models matched to this evaluation: Ranks #4 of 102 on LMArena's WebDev coding arena (Elo 1768), a leaderboard built from blind human preference votes on coding tasks.”
    LMArena WebDev (coding) arenaOpen original ↗

    No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.

  5. 05

    Kimi K3.1 ProMoonshot AI · $0.24 in / $8.40 out per M tokens · 1.0M context · open-weight

    Kimi K3.1 Pro ranks #7 of 120 on Design Arena's coding category (Elo 1350), based on blind human preference between generated coding results.

    Best when: Its strongest result for this task is on Design Arena coding, which measures coding results preferred in blind head-to-head comparisons.
    Source 3
    “Within the 120 catalog models matched to this evaluation: Ranks #7 of 120 on Design Arena's coding category (Elo 1350), based on blind human preference between generated coding results.”
    Design Arena codingOpen original ↗

    No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.

  6. 06

    Gemini 4 HydrogenGoogle · $1.50 in / $7.50 out per M tokens · 1.0M context

    Gemini 4 Hydrogen ranks #8 of 102 on LMArena's WebDev coding arena (Elo 1704), a leaderboard built from blind human preference votes on coding tasks.

    Best when: Its strongest result for this task is on LMArena WebDev, which measures practical, real-world coding quality.
    Source 6
    “Within the 102 catalog models matched to this evaluation: Ranks #8 of 102 on LMArena's WebDev coding arena (Elo 1704), a leaderboard built from blind human preference votes on coding tasks.”
    LMArena WebDev (coding) arenaOpen original ↗

    No practitioner reports are linked yet, so no cautions are listed. Absence of reports is not evidence of reliability.

  7. 07

    Gemini 4 ArgonGoogle · $2 in / $10 out per M tokens · 1.0M context

    Ranks #12 on Design Arena coding category (Elo 1326) and #11 on LMArena WebDev coding arena (Elo 1681), with 55.45% on Terminal-Bench 4.1.

    Best when: Use for TUI (terminal user interface) development where practitioners report strong performance.
    Source 7
    “Same issue on opencode linux, problem is certainly in the endpoint, not the client. Sad, Gemini Argon was really good at making TUIs”

    Watch out for

    • Terminal task performance of 55.45% ranks #10 of 46 on Terminal-Bench 4.1.
      Source 8
      “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-BenchOpen original ↗
  8. 08

    GPT-6.1 SolOpenAI · $2 in / $10 out per M tokens · 1.1M context

    GPT-6.1 Sol ranks #4 of 120 on Design Arena's coding category (Elo 1362), based on blind human preference between generated coding results.

    Best when: Its strongest result for this task is on Design Arena coding, which measures coding results preferred in blind head-to-head comparisons.
    Source 9
    “Within the 120 catalog models matched to this evaluation: Ranks #4 of 120 on Design Arena's coding category (Elo 1362), based on blind human preference between generated coding results.”
    Design Arena codingOpen original ↗

Freshness check

New models awaiting independent results

These recently released, tool-capable models are relevant to coding but do not yet have enough independent evidence to place confidently in the ranking above.

Frequently asked

What is the best model for Coding?
Claude Fable 5.2 ranks first as of November 15, 2026. Claude Fable 5.2 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.[1]
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.[2]
Which open-weight model ranks highest for Coding?
Kimi K3.1 Pro is the highest-ranked open-weight model, at #5. Kimi K3.1 Pro ranks #7 of 120 on Design Arena's coding category (Elo 1350), based on blind human preference between generated coding results.[3]

Sources

  1. 1

    “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
  2. 2

    “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
  3. 3

    “Within the 120 catalog models matched to this evaluation: Ranks #7 of 120 on Design Arena's coding category (Elo 1350), based on blind human preference between generated coding results.”

    Design Arena coding · Benchmark · Nov 15, 2026
  4. 4

    “Within the 59 catalog models matched to this evaluation: Scores 91.37% on LiveBench Coding (#1 of 59), an objective evaluation of code generation and completion.”

    LiveBench Coding · Benchmark · Oct 18, 2026
  5. 5

    “Within the 102 catalog models matched to this evaluation: Ranks #4 of 102 on LMArena's WebDev coding arena (Elo 1768), a leaderboard built from blind human preference votes on coding tasks.”

    LMArena WebDev (coding) arena · Benchmark · Nov 11, 2026
  6. 6

    “Within the 102 catalog models matched to this evaluation: Ranks #8 of 102 on LMArena's WebDev coding arena (Elo 1704), a leaderboard built from blind human preference votes on coding tasks.”

    LMArena WebDev (coding) arena · Benchmark · Nov 11, 2026
  7. 7

    “Same issue on opencode linux, problem is certainly in the endpoint, not the client. Sad, Gemini Argon was really good at making TUIs”

    loeclos · GitHub · Nov 5, 2026
  8. 8

    “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
  9. 9

    “Within the 120 catalog models matched to this evaluation: Ranks #4 of 120 on Design Arena's coding category (Elo 1362), based on blind human preference between generated coding results.”

    Design Arena coding · Benchmark · Nov 15, 2026

Rankings synthesized from community evidence and open benchmarks. See methodology. Not driven by vendor marketing.