SYLIN

Claude Opus

by Anthropic · Model · Language & code models

strong

Still a top-three model for agents and coding, now behind Fable 5.2.

MAC Bench finds most gains from eight agents came from parallel exploration; Opus 5.5 was the only tested model with extra gain from coordination

Why we say so

Arena ranks Opus 5.5 second on agents and web development, and Artificial Analysis ties Fable 5.2 with it on its coding agent index. Bespoke Labs found it the only tested model that gained from coordinating multiple agents. HN commenters report its safeguards block biology and DNA work about half the time, and Gemini Neon is said to match or slightly beat it on Vals.

Versions

VersionReleasedRank
Claude Opus 5.5LATEST
strong Second on Arena agents and web development, with strong coding scores
Sep 24#2 image understanding#2 web development#3 agents
Claude Opus 5Jul 25#2 documents
Claude Opus 4.8
mixed Usable for reverse engineering but needs babysitting on harder targets
Claude Opus 4.7
Claude Opus 4.6#3 search
Claude Opus 4

Evidence

  1. Nov 19
    Main result first obtained by an internal Claude research model; public Opus 5.5 later reproduced it independently, per the Hexagon entry
  2. Nov 18
    Predicts the next amino acid of unseen proteins: 34.9% real vs 7.5% shuffled, a +27.4 point gap, suggesting protein-sequence training
  3. Nov 18
    Trails ProGen3-3B (+32.5 gap) on next-residue prediction
  4. Nov 18
    Roughly matches ProtGPT3-10B (+27.7 gap) on next-residue prediction
  5. Nov 15
    #2 on Arena agent
Show all 46
  1. Nov 15
    #2 on Arena webdev (1818)
  2. Nov 12
    Opus 5.5 at xhigh outscores Sonnet 5.5 at max with 40% lower cost per task: a higher tier beats more effort
  3. Nov 11
    #2 on Arena webdev (1815)
  4. Nov 11
    #3 on Arena text (1543)
  5. Nov 9
    #2 on Arena document (1509)
  6. Nov 9
    #3 on Arena search (1252)
  7. Nov 9
    #2 on Arena vision (1324)
  8. Oct 19
    #3 on Arena search (1251)
  9. Oct 11
    Said to be even better than GPT-6 Astra at reverse engineering binaries; no figures given
  10. Oct 8
    Took #1 on Arena agent
  11. Sep 27
    Claude Opus 5.5 released; frontier-level at 3D, design, coding and UI navigation but pricier than GPT-6 Astra
  12. Sep 27
    Presenter preferred Opus on piano, math video, Blender and Unreal tests; Astra on the water-balloon sim
  13. Sep 24
    Self-reported results beat Fable 5.1 and GPT-6 Astra on agentic coding, knowledge work and computer use
  14. Sep 24
    Ranked first on Vals Index; GPT-6 Astra fourth
  15. Sep 24
    Ranked first on KernelBench, well ahead of GPT-6 Astra
  16. Sep 24
    Lowest hallucination rate among Claude models but higher than GPT-6, Grok and Muse; about $6 per task, above Astra
  17. Sep 24
    Got 1 of 6 brain-tumor types right, tied with Astra, GLM 5.3 and Chimera 3 as the best so far
  18. Sep 24
    Fable-style guardrails: risky cyber, biology or distillation requests fall back to a weaker model
  19. Sep 23
    Took #1 on Arena webdev
  20. Sep 13
    One commenter finds Opus 5 more verbose and harder to read than Opus 4.6 and 4.8, which they prefer
  21. Sep 4
    Designed a textbook 74-series and GAL circuit for 640x480 VGA output from EEPROM; one uncaught error, fixed with a blue wire, then worked
  22. Sep 4
    Tops the 1 Sept EEBench leaderboard at 61.6%, ahead of Grok 4.6 (57.1%) and Claude Fable 5.1 (56.4%)
  23. Aug 31
    Model reliably reaches for `python -c` decoders, making it exploitable via Python's cwd-first import path
  24. Aug 28
    Top of Terminal-Bench-Science at 30.0% resolution with Claude Code, ahead of GPT Sol 22.4% and Fable 21.4%
  25. Aug 22
    Opus 5 (max effort) ranked second at 2,920 (53.6% gap closed) in 2.9 agent-days, ahead of Kimi K3 and GPT Sol
  26. Aug 21
    Commenters report Opus 5 is slower (a multi-step skill took 2 hours versus 20 minutes on older models) and worse than 4.8
  27. Aug 20
    Users report jargon-heavy, over-compressed prose; 'load-bearing' appears hundreds of times in one codebase's output
  28. Jul 27
    Anthropic's Claude Opus 5 claimed stronger reasoning and planning than Fable 5 on agentic coding; on all paid plans and API
  29. Jul 27
    Researched Q4 reports, then built a one-minute narrated financial video with HyperFrames and Gemini TTS from one prompt, self-verifying; cleaner than Kimi K3
  30. Jul 27
    Kebab-skewer HTML prompt took about 10 minutes and roughly $6-7; output beat Fable 5 and Opus 4.8 on flames and atmosphere
  31. Jul 27
    On DeepSWE, GPT Sol still ranks first; Anthropic's chart de-emphasised it. Opus 5 beats Fable on agentic coding but trails on legal and health
  32. Jul 27
    Ranks 2nd on Simple Bench behind Fable, 3rd on LiveBench behind GPT Sol and Fable, and 2nd on Vals Index just ahead of Kimi K3
  33. Jul 27
    From one prompt in Claude Code it built a working browser Windows 11 clone with Word, Excel, Store and games; PowerPoint was weak
  34. Jul 27
    Got 0 of 6 tumor scans right, while Chimera 3 got one; unlike Fable 5 it did not refuse the request
  35. Jul 27
    Strong at front end and 3D design, but roughly a tie with GPT 5.6 and Kimi K3 elsewhere; about twice as slow, much pricier
  36. Jul 27
    Anthropic reports over 30% on ARC-AGI-3, far above the next best model, GPT-5.6, at under 10%
  37. Jul 27
    Ranked first on Artificial Analysis Intelligence Index by one point over Fable 5, with no confidence intervals shown
  38. Jul 25
    Same kebab prompt: Opus 5 slightly ahead of Kimi K3 on flames and photorealism, Kimi's meat texture arguably better; subjective
  39. Mar 31
    Took #1 on Arena vision
  40. Mar 3
    Took #1 on Arena document
  41. Feb 25
    Took #1 on Arena search

Cost to run

Alternatives

Claude Fable#1 on Arena agent
GPT Sol#2 on Arena search (1258)
GPT AstraBeats Claude Fable 5.2 on Agents' Last Exam at much lower cost
GLMIntelligence index score within one or two points of the best GPT and Claude models
KimiKimi K3 open weights reached 2,930 (52.2% gap closed) with Prime Agent harness; 2,974 with kimi-code, near Opus 5
DeepSeek FlashNarrated Earth-circumference explainer video (36 min, HyperFrames, Gemini TTS) matched frontier model quality