Chat turn cost
1K fresh input + 500 output tokens
Claude Opus 5
Claude Opus 5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Claude Fable 5.2 vs Claude Opus 5
Updated November 9, 2026. Rank says Claude Fable 5.2 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.
Share or export
Decision reading
Claude Fable 5.2 has the higher public point estimate, 91.19 versus 79.11. Their conditional score ranges overlap. These ranges do not establish rank confidence. 19 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 19
- Claude Fable 5.2 only
- 11
- Claude Opus 5 only
- 65
- Like-for-like categories
- 2 / 8
Estimated: Claude Fable 5.2 · Supported: Claude Opus 5. Conditional ranges do not establish rank confidence.How the comparison works
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Claude Opus 5
Claude Opus 5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Show secondary and unsupported calls
Repository review cost
50K fresh input + 3K output tokens
Claude Opus 5
Claude Opus 5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-ratesCoding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Claude Fable 5.2 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limitedAgentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Claude Fable 5.2 is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limitedLong documents
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Confidence: documented
Which one for a specific job
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
86.9Claude Fable 5.270.4Claude Opus 5
Directional only · BenchAlign v5.8
Claude Fable 5.2 has the higher coding point estimate. Conditional score ranges do not establish rank confidence.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Same basis rules as the category table below
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
11 Claude Fable 5.2 only
19 Shared
65 Claude Opus 5 only
3 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
ARC-AGI-3Reasoning
Normalized gap 35.5Claude Fable 5.2:65.7%Claude Opus 5:30.2%AutomationBenchAgentic
Normalized gap 25.4Claude Fable 5.2:51.4%Claude Opus 5:26.0%HLE w/o toolsKnowledge
Normalized gap 16.9Claude Fable 5.2:73.2%Claude Opus 5:56.3%SWE-bench ProCoding
Normalized gap 13.8Claude Fable 5.2:93%Claude Opus 5:79.2%HLEKnowledge
Normalized gap 9.5Claude Fable 5.2:74.2%Claude Opus 5:64.7%
Bars run 0–100 on each benchmark’s normalized display scale
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, reasoning, multimodal, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
Reasoning
Like-for-like- Claude Fable 5.2
- 90.7
- Supported · #1/157
- Claude Opus 5
- 74.8
- Supported · #11/157
- Basis
- BenchAlign v5.8 lane · 3 vs 3 public rows
- Reading
- Claude Fable 5.2 leads
Knowledge
Like-for-like- Claude Fable 5.2
- 97.6
- Supported · #1/179
- Claude Opus 5
- 80.3
- Supported · #6/179
- Basis
- BenchAlign v5.8 lane · 3 vs 19 public rows
- Reading
- Claude Fable 5.2 leads
Agentic
Directional only- Claude Fable 5.2
- 90.7
- Estimated · #1/132
- Claude Opus 5
- 77.0
- Supported · #5/132
- Basis
- BenchAlign v5.8 lane · 8 vs 23 public rows
- Reading
- Directional only
Coding
Directional only- Claude Fable 5.2
- 86.9
- Estimated · #1/158
- Claude Opus 5
- 70.4
- Supported · #10/158
- Basis
- BenchAlign v5.8 lane · 8 vs 20 public rows
- Reading
- Directional only
Multimodal
Directional only- Claude Fable 5.2
- 89.8
- Estimated · #1/67
- Claude Opus 5
- 77.8
- Estimated · #6/67
- Basis
- BenchAlign v5.8 lane · 6 vs 8 public rows
- Reading
- Directional only
Multilingual
Not comparable- Claude Fable 5.2
- Not ranked
- Claude Opus 5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Claude Fable 5.2
- Not ranked
- Claude Opus 5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Claude Fable 5.2
- Not ranked
- Claude Opus 5
- 88.1
- Unranked · 7 rankable rows
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Not comparable
| Category | Claude Fable 5.2 | Claude Opus 5 | Basis | Reading |
|---|---|---|---|---|
| Reasoning | 90.7Supported · #1/157 | 74.8Supported · #11/157 | Like-for-likeBenchAlign v5.8 lane · 3 vs 3 public rows | Claude Fable 5.2 leads |
| Knowledge | 97.6Supported · #1/179 | 80.3Supported · #6/179 | Like-for-likeBenchAlign v5.8 lane · 3 vs 19 public rows | Claude Fable 5.2 leads |
| Agentic | 90.7Estimated · #1/132 | 77.0Supported · #5/132 | Directional onlyBenchAlign v5.8 lane · 8 vs 23 public rows | Directional only |
| Coding | 86.9Estimated · #1/158 | 70.4Supported · #10/158 | Directional onlyBenchAlign v5.8 lane · 8 vs 20 public rows | Directional only |
| Multimodal | 89.8Estimated · #1/67 | 77.8Estimated · #6/67 | Directional onlyBenchAlign v5.8 lane · 6 vs 8 public rows | Directional only |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | Not ranked | 88.1Unranked · 7 rankable rows | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, reasoning, multimodal, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Supported evidence per lane · bars run 0–100Methodology
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- Claude Fable 5.2
- $0.028
- Fits in one request
- Claude Opus 5
- $0.0175
- Fits in one request
Claude Opus 5 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Claude Fable 5.2
- $0.52
- Fits in one request
- Claude Opus 5
- $0.325
- Fits in one request
Claude Opus 5 has the lower modeled cost
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Claude Fable 5.2
- $0.6
- Fits in one request
- Claude Opus 5
- $0.45
- Fits in one request
Claude Opus 5 has the lower modeled cost
Costs use the listed standard API rates.
Cached input falls back to the list input rate only where a cached rate is unpublished
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationClaude Fable 5.2Claude Opus 5
Context window
Maximum documented context; output-token limits may be lower.
Claude Fable 5.2
Claude Opus 5
API model ID
Claude Fable 5.2
claude-fable-5-2
Anthropic Claude Fable 5.2 model documentationClaude Opus 5
claude-opus-5
Anthropic model overviewCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Claude Fable 5.2
$0.2 per 1M cached input tokens
Claude API pricingClaude Opus 5
$0.5 per 1M cached input tokens
Claude API pricingDocumented inputs
Claude Fable 5.2
Claude Opus 5
text, image
Anthropic model overviewDocumented outputs
Claude Fable 5.2
Claude Opus 5
Provider availability
Claude Fable 5.2
Generally Available · Claude API, Claude products, AWS, Google Cloud, Microsoft Azure
Anthropic Fable 5.2 and Mythos 5.2 launchClaude Opus 5
Generally Available · Claude API
Anthropic model overviewReasoning profile
Claude Fable 5.2
Reasoning
Claude Opus 5
Reasoning
Weight access
Claude Fable 5.2
Proprietary
Claude Opus 5
Proprietary
License
Claude Fable 5.2
Proprietary
Claude Opus 5
Proprietary
Release date
Claude Fable 5.2
2026-11-09
Claude Opus 5
2026-07-24
If you already use one of these models
- Deployment change
- Both entries list Anthropic as the provider. Confirm endpoint, model ID, limits, and feature support before switching.
- Quality signal
- Claude Fable 5.2 has the higher public point estimate, 91.19 versus 79.11. Their conditional score ranges overlap. These ranges do not establish rank confidence.
- Workload cost
- Repository review: $0.52 vs $0.325. Cache-heavy agent loop: $0.6 vs $0.45.
- Context tradeoff
- Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Claude Fable 5.2 or Claude Opus 5?
Claude Fable 5.2 has the higher public point estimate, 91.19 versus 79.11. Their conditional score ranges overlap. These ranges do not establish rank confidence. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, Claude Fable 5.2 or Claude Opus 5?
Claude Fable 5.2 scores higher for coding on the public lane, 86.9 to 70.4. Claude Fable 5.2 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which is better for agentic tasks, Claude Fable 5.2 or Claude Opus 5?
Claude Fable 5.2 scores higher for agentic tasks on the public lane, 90.7 to 77. Claude Fable 5.2 is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which costs less, Claude Fable 5.2 or Claude Opus 5?
For the stated presets, chat costs $0.028 on Claude Fable 5.2 and $0.0175 on Claude Opus 5; repository review costs $0.52 and $0.325; the cache-heavy agent loop costs $0.6 and $0.45. Costs use the listed standard API rates.
Which has the larger context window, Claude Fable 5.2 or Claude Opus 5?
Both models list the same context window, 1M.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence95 rows
Agentic
Terminal-Bench 4.0
Claude Fable 5.267.60%SourceClaude Opus 5—Not directly comparable
Terminal-Bench-Science 0.1
Claude Fable 5.274.6%SourceClaude Opus 5—Not directly comparable
AutomationBench
Claude Fable 5.251.4%SourceClaude Opus 526.0%SourceClaude Fable 5.2 leads this result
OSWorld 2.1 Offline
Claude Fable 5.289.3%SourceClaude Opus 5—Not directly comparable
WANDR
Claude Fable 5.282.1%SourceClaude Opus 5—Not directly comparable
Remote Labor Index
Claude Fable 5.241.3%SourceClaude Opus 5—Not directly comparable
RoboDojo Sim
Claude Fable 5.264.9%SourceClaude Opus 5—Not directly comparable
DRACO
Claude Fable 5.287.7%SourceClaude Opus 588.6%SourceClaude Opus 5 leads this result
Terminal-Bench 4.1
Claude Fable 5.2—Claude Opus 553.30%SourceNot directly comparable
Terminal-Bench 3.0
Claude Fable 5.2—Claude Opus 542.7%SourceNot directly comparable
BrowseComp
Claude Fable 5.2—Claude Opus 590.8%SourceNot directly comparable
HLE w/ tools
Claude Fable 5.2—Claude Opus 564.7%SourceNot directly comparable
DeepSearchQA
Claude Fable 5.2—Claude Opus 595.0%SourceNot directly comparable
BrowseComp (10-agent, prerelease)
Claude Fable 5.2—Claude Opus 593.6%SourceNot directly comparable
OSWorld 2.0
Claude Fable 5.2—Claude Opus 570.6%SourceNot directly comparable
MCP Atlas
Claude Fable 5.2—Claude Opus 585.8%SourceNot directly comparable
MCP-Atlas claim coverage
Claude Fable 5.2—Claude Opus 589.1%SourceNot directly comparable
LAB all-pass (Anthropic harness)
Claude Fable 5.2—Claude Opus 523.58%SourceNot directly comparable
LAB criterion-pass (Anthropic harness)
Claude Fable 5.2—Claude Opus 593.74%SourceNot directly comparable
LAB all-pass (Harvey held-out)
Claude Fable 5.2—Claude Opus 511.7%SourceNot directly comparable
LAB criterion-pass (Harvey held-out)
Claude Fable 5.2—Claude Opus 594.1%SourceNot directly comparable
Toolathlon-Verified
Claude Fable 5.2—Claude Opus 580.6%SourceNot directly comparable
Toolathlon Verified Pass@3
Claude Fable 5.2—Claude Opus 587.0%SourceNot directly comparable
Toolathlon Verified Pass³
Claude Fable 5.2—Claude Opus 573.1%SourceNot directly comparable
Toolathlon Verified avg. turns
Claude Fable 5.2—Claude Opus 523.5 turnsSourceNot directly comparable
Terminal-Bench 2.1 (Vals)
Claude Fable 5.2—Claude Opus 584.6%SourceNot directly comparable
ApprenticeBench
Claude Fable 5.2—Claude Opus 536%SourceNot directly comparable
MazeBench
Claude Fable 5.2—Claude Opus 512%SourceNot directly comparable
OSWorld 2.1
Claude Fable 5.2—Claude Opus 574.0%SourceNot directly comparable
Coding
SWE-bench Pro
Claude Fable 5.293%SourceClaude Opus 579.2%SourceClaude Fable 5.2 leads this result
SWE Multilingual
Claude Fable 5.296.6%SourceClaude Opus 589.5%SourceClaude Fable 5.2 leads this result
SWE Multimodal
Claude Fable 5.274.1%SourceClaude Opus 559.4%SourceClaude Fable 5.2 leads this result
DeepSWE
Claude Fable 5.274.3%SourceClaude Opus 568.8%SourceClaude Fable 5.2 leads this result
FrontierSWE v2
Shared sourceClaude Fable 5.269.9%Claude Opus 552.0%Claude Fable 5.2 leads this result
ProgramBench
Claude Fable 5.285.6%SourceClaude Opus 593.0%SourceClaude Opus 5 leads this result
FrontierCode 2.0
Shared sourceClaude Fable 5.249.1%Claude Opus 526.7%Claude Fable 5.2 leads this result
CursorBench 4.0
Shared sourceClaude Fable 5.260.0%Claude Opus 546.6%Claude Fable 5.2 leads this result
ScarfBench
Claude Fable 5.2—Claude Opus 569.8%SourceNot directly comparable
Bug Hunt Bench
Claude Fable 5.2—Claude Opus 527.0 fixesSourceNot directly comparable
SWE-bench Verified
Claude Fable 5.2—Claude Opus 596%SourceNot directly comparable
FrontierCode 1.1 Main
Claude Fable 5.2—Claude Opus 553.4%SourceNot directly comparable
FrontierCode 1.1 Extended
Claude Fable 5.2—Claude Opus 563.6%SourceNot directly comparable
ProgramBench (episode 1)
Claude Fable 5.2—Claude Opus 583.0%SourceNot directly comparable
CursorBench 3.2
Claude Fable 5.2—Claude Opus 570.0%SourceNot directly comparable
VulcanBench v3
Claude Fable 5.2—Claude Opus 587.0%SourceNot directly comparable
VulcanBench CII v1
Claude Fable 5.2—Claude Opus 596.4%SourceNot directly comparable
LiveCodeBench (Vals)
Claude Fable 5.2—Claude Opus 589.0%SourceNot directly comparable
SWE-bench (Vals)
Claude Fable 5.2—Claude Opus 597.0%SourceNot directly comparable
PostTrainBench v1.1
Claude Fable 5.2—Claude Opus 535.0%SourceNot directly comparable
Reasoning
ARC-AGI-3
Claude Fable 5.265.7%SourceClaude Opus 530.2%SourceClaude Fable 5.2 leads this result
ARC-AGI-3 Provider Adapter
Claude Fable 5.294.90%SourceClaude Opus 5—Not directly comparable
STEM Professor Bench
Claude Fable 5.261.8%SourceClaude Opus 5—Not directly comparable
ARC-AGI-1
Claude Fable 5.2—Claude Opus 597.50%SourceNot directly comparable
ARC-AGI-2
Claude Fable 5.2—Claude Opus 590.4%SourceNot directly comparable
Multimodal
Chartography (no tools)
Claude Fable 5.273.4%SourceClaude Opus 529.6%SourceClaude Fable 5.2 leads this result
Chartography (tools)
Claude Fable 5.291.6%SourceClaude Opus 583.0%SourceClaude Fable 5.2 leads this result
OfficeQA
Claude Fable 5.283.4%SourceClaude Opus 578.1%SourceClaude Fable 5.2 leads this result
OfficeQA Pro
Claude Fable 5.272.9%SourceClaude Opus 566.9%SourceClaude Fable 5.2 leads this result
BenchCAD 2.1
Claude Fable 5.275.4%SourceClaude Opus 5—Not directly comparable
BenchCAD 2.1 (tools)
Claude Fable 5.283.5%SourceClaude Opus 5—Not directly comparable
BenchCAD Vision2Code (no tools)
Claude Fable 5.2—Claude Opus 50.366SourceNot directly comparable
BenchCAD Vision2Code (tools)
Claude Fable 5.2—Claude Opus 50.821SourceNot directly comparable
GDP.pdf (no tools)
Claude Fable 5.2—Claude Opus 583.4%SourceNot directly comparable
GDP.pdf (tools)
Claude Fable 5.2—Claude Opus 585.5%SourceNot directly comparable
Knowledge
HLE
Claude Fable 5.274.2%SourceClaude Opus 564.7%SourceClaude Fable 5.2 leads this result
HLE w/o tools
Claude Fable 5.273.2%SourceClaude Opus 556.3%SourceClaude Fable 5.2 leads this result
PhysicianBench
Claude Fable 5.264.4%SourceClaude Opus 5—Not directly comparable
HLE-Verified
Claude Fable 5.2—Claude Opus 554.4%SourceNot directly comparable
LABBench2
Claude Fable 5.2—Claude Opus 584.2%SourceNot directly comparable
HealthBench (raw)
Claude Fable 5.2—Claude Opus 567.1%SourceNot directly comparable
HealthBench (length-adjusted)
Claude Fable 5.2—Claude Opus 557.8%SourceNot directly comparable
HealthBench Professional
Claude Fable 5.2—Claude Opus 559.8%SourceNot directly comparable
HealthBench Professional (raw)
Claude Fable 5.2—Claude Opus 573.4%SourceNot directly comparable
BioMysteryBench (human-solvable)
Claude Fable 5.2—Claude Opus 590.1%SourceNot directly comparable
BioMysteryBench (human-difficult)
Claude Fable 5.2—Claude Opus 549.4%SourceNot directly comparable
SpatialBench Verified
Claude Fable 5.2—Claude Opus 572.5%SourceNot directly comparable
SingleCellBench
Claude Fable 5.2—Claude Opus 560.6%SourceNot directly comparable
ProteinGym Hard
Claude Fable 5.2—Claude Opus 547.7%SourceNot directly comparable
Protein Design
Claude Fable 5.2—Claude Opus 542.5%SourceNot directly comparable
Organic chemistry V2
Claude Fable 5.2—Claude Opus 561.6%SourceNot directly comparable
Protocols (troubleshooting)
Claude Fable 5.2—Claude Opus 561.1%SourceNot directly comparable
Protocols (understanding)
Claude Fable 5.2—Claude Opus 578.4%SourceNot directly comparable
GPQA Diamond (Vals)
Claude Fable 5.2—Claude Opus 593.4%SourceNot directly comparable
MMLU-Pro (Vals)
Claude Fable 5.2—Claude Opus 591.6%SourceNot directly comparable
Multilingual
Instruction following
Gray Swan IPI (15 attempts)
Claude Fable 5.2—Claude Opus 54.6%SourceNot directly comparable
Math
IMO 2026
Claude Fable 5.2—Claude Opus 542/42SourceNot directly comparable
RiemannBench (no tools)
Claude Fable 5.2—Claude Opus 560.0%SourceNot directly comparable
RiemannBench (tools)
Claude Fable 5.2—Claude Opus 579.0%SourceNot directly comparable
ArXivMath Jun. 2026 (no tools)
Claude Fable 5.2—Claude Opus 590.8%SourceNot directly comparable
ArXivMath Jun. 2026 (tools)
Claude Fable 5.2—Claude Opus 591.3%SourceNot directly comparable
FrontierMath v2 (Tiers 1-3)
Claude Fable 5.2—Claude Opus 585.614%SourceNot directly comparable
FrontierMath v2 (Tier 4)
Claude Fable 5.2—Claude Opus 573.171%SourceNot directly comparable
95 public results · 19 shared
Watch Claude Fable 5.2 vs Claude Opus 5
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 6,100+ subscribers.
Last updated November 9, 2026