Agentic
Directional only- BTL-4
- 50.6
- Estimated · #50/152
- Claude Sonnet 5
- 65.8
- Supported · #11/152
- Basis
- BenchAlign lane · 1 vs 6 public rows
- Reading
- Directional only
BTL-4 vs Claude Sonnet 5
Updated September 8, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
1 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Share or export
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.
Long documents
Prompts that approach the documented context limit
Claude Sonnet 5
Claude Sonnet 5 has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
BTL-4 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
BTL-4 is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
2 BTL-4 only
1 Shared
23 Claude Sonnet 5 only
- Shared results
- 1
- BTL-4 only
- 2
- Claude Sonnet 5 only
- 23
- Like-for-like categories
- 0 / 8
2 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, 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.
Coding
Directional only- BTL-4
- 50.5
- Estimated · #60/151
- Claude Sonnet 5
- 64.1
- Supported · #13/151
- Basis
- BenchAlign lane · 2 vs 10 public rows
- Reading
- Directional only
Reasoning
Not comparable- BTL-4
- Not ranked
- Claude Sonnet 5
- 77.4
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- BTL-4
- Not ranked
- Claude Sonnet 5
- 66.7
- Supported · #20/182
- Basis
- BenchAlign lane · 0 vs 6 public rows
- Reading
- Not comparable
Math
Not comparable- BTL-4
- Not ranked
- Claude Sonnet 5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- BTL-4
- Not ranked
- Claude Sonnet 5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- BTL-4
- Not ranked
- Claude Sonnet 5
- 77.5
- #13/48
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- BTL-4
- Not ranked
- Claude Sonnet 5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | BTL-4 | Claude Sonnet 5 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 50.6Estimated · #50/152 | 65.8Supported · #11/152 | Directional onlyBenchAlign lane · 1 vs 6 public rows | Directional only |
| Coding | 50.5Estimated · #60/151 | 64.1Supported · #13/151 | Directional onlyBenchAlign lane · 2 vs 10 public rows | Directional only |
| Reasoning | Not ranked | 77.4Unranked · 2 rankable rows | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Knowledge | Not ranked | 66.7Supported · #20/182 | Not comparableBenchAlign lane · 0 vs 6 public rows | Not comparable |
| Math | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | Not ranked | 77.5#13/48 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Instruction following | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
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.
SWE-bench Verified
Coding
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
- BTL-4
- Self-hosted; infrastructure cost varies
- Fits in one request
- Claude Sonnet 5
- $0.007
- Fits in one request
BTL-4 has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- BTL-4
- Self-hosted; infrastructure cost varies
- Fits in one request
- Claude Sonnet 5
- $0.13
- Fits in one request
BTL-4 has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- BTL-4
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
- Claude Sonnet 5
- $0.18
- Fits in one request
BTL-4 has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationBTL-4Claude Sonnet 5
Context window
Maximum documented context; output-token limits may be lower.
BTL-4
Claude Sonnet 5
API model ID
BTL-4
Not sourced
Claude Sonnet 5
claude-sonnet-5
Anthropic model overviewCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
BTL-4
No comparable hosted API rate
Bad Theory Labs BTL-4 model cardClaude Sonnet 5
$0.2 per 1M cached input tokens
Claude API pricingDocumented inputs
BTL-4
Not sourced
Claude Sonnet 5
text, image
Anthropic model overviewDocumented outputs
BTL-4
Not sourced
Claude Sonnet 5
Provider availability
BTL-4
Not sourced
Claude Sonnet 5
Generally Available · Claude API
Anthropic model overviewReasoning profile
BTL-4
Reasoning
Claude Sonnet 5
Reasoning
Weight access
BTL-4
Open Weight
Claude Sonnet 5
Proprietary
License
BTL-4
Open Weight
Claude Sonnet 5
Proprietary
Release date
BTL-4
2026-08-05
Claude Sonnet 5
2026-06-30
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Claude Sonnet 5 has the larger documented window (1M).
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence26 rows
Agentic
BFCL v4
BTL-473.5%SourceClaude Sonnet 5—Not directly comparable
Terminal-Bench 3.0
BTL-4—Claude Sonnet 514.6%SourceNot directly comparable
Terminal-Bench 2.0
BTL-4—Claude Sonnet 580.4%SourceNot directly comparable
BrowseComp
BTL-4—Claude Sonnet 584.7%SourceNot directly comparable
HLE w/ tools
BTL-4—Claude Sonnet 557.4%SourceNot directly comparable
OSWorld-Verified
BTL-4—Claude Sonnet 581.2%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
BTL-4—Claude Sonnet 574.5%SourceNot directly comparable
Coding
SWE-bench Verified
BTL-478.4%SourceClaude Sonnet 585.2%SourceClaude Sonnet 5 leads this result
LiveCodeBench v6
BTL-466.1%SourceClaude Sonnet 5—Not directly comparable
SWE-bench Pro
BTL-4—Claude Sonnet 563.2%SourceNot directly comparable
SWE Multilingual
BTL-4—Claude Sonnet 578.3%SourceNot directly comparable
SWE Multimodal
BTL-4—Claude Sonnet 528.1%SourceNot directly comparable
Terminal-Bench 2.0
BTL-4—Claude Sonnet 580.4%SourceNot directly comparable
FrontierCode 1.1 Main
BTL-4—Claude Sonnet 542.7%SourceNot directly comparable
cursorBench32
BTL-4—Claude Sonnet 561.5%SourceNot directly comparable
VulcanBench CII v1
BTL-4—Claude Sonnet 589.2%SourceNot directly comparable
LiveCodeBench (Vals)
BTL-4—Claude Sonnet 582.4%SourceNot directly comparable
SWE-bench (Vals)
BTL-4—Claude Sonnet 579.6%SourceNot directly comparable
Knowledge
HLE
BTL-4—Claude Sonnet 557.4%SourceNot directly comparable
HLE w/o tools
BTL-4—Claude Sonnet 543.2%SourceNot directly comparable
HLE-Verified
BTL-4—Claude Sonnet 531.0%SourceNot directly comparable
LABBench2
BTL-4—Claude Sonnet 580.1%SourceNot directly comparable
GPQA Diamond (Vals)
BTL-4—Claude Sonnet 588.9%SourceNot directly comparable
MMLU-Pro (Vals)
BTL-4—Claude Sonnet 587.5%SourceNot directly comparable
Multimodal
Frequently asked questions
Which is better, BTL-4 or Claude Sonnet 5?
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, BTL-4 or Claude Sonnet 5?
Claude Sonnet 5 scores higher for coding on the public lane, 64.1 to 50.5. BTL-4 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, BTL-4 or Claude Sonnet 5?
Claude Sonnet 5 scores higher for agentic tasks on the public lane, 65.8 to 50.6. BTL-4 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, BTL-4 or Claude Sonnet 5?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, BTL-4 or Claude Sonnet 5?
Claude Sonnet 5 has the larger documented context window: 1M, compared with 262K.
Related comparisons
Last updated September 8, 2026
Watch BTL-4 vs Claude Sonnet 5
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.