BTL-4 released
Bad Theory Labs · Model release
Source confirmed
Data as of September 24, 2026 · How the score is built
CurrentOpen WeightReasoning
Released Aug 5, 2026262K contextBad Theory Labs BTL-4 model card
BTL-4
Decision readingBTL-4 is tracked, but not publicly ranked yet. The profile exposes 3 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
Released Aug 5, 2026 — see all recent releases
BTL-4 will be repriced, updated or retired. Get each notice with its source and date. Follow model changes
Share or export
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
Unranked
field median 50.2Not eligible for a public rank
Price
Self-hosted; infrastructure cost varies
input median $0.97No comparable first-party hosted token rate
Speed
Not measured
field median 91 tok/sTime to first token not measured
Context
262Ktokens
field median 256,000Maximum output length is tracked separately
Strongest published evidence
Published rows are visible, but no category has enough eligible evidence for a comparative rank.
Validate before choosing
3 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.
Source-linked · 3 displayable benchmark rows
Follow model changesCategory score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticWeight 22%1 benchmarkVerified | Score pending | 1 benchmark | Verified | |||
| CodingRank Not rankedWeight 20%2 benchmarksVerified | 61.5 | 2 benchmarks | Verified | |||
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| KnowledgeWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MathWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured |
3 of 483 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic1/1 verified
- Coding2/2 verified
- ReasoningNot measured
- MultimodalNot measured
- KnowledgeNot measured
- MultilingualNot measured
- Inst. FollowingNot measured
- MathNot measured
Verified sourceProvisionalNot measured
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | Score78.4% | Versus best verified row Best verified: Claude Opus 5 · 96% | Gap17.6 behind | WeightDisplay only | Provider exact |
| LiveCodeBench v6 | Score66.1% | Versus best verified row Best verified: Sakana Fugu-Ultra · 93.2% | Gap27.1 behind | WeightDisplay only | Provider exact |
Agentic1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| BFCL v4Berkeley Function Calling Leaderboard v4 | Score73.5% | Versus best verified row Best verified: BTL-3 · 88.5% | Gap15 behind | Weight3% ref. weight | Provider exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 3 rowsLineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Radar
Full release historyBTL-4 release history
Radar confirmed these at the source. Use BTL-4 in your work? Explore Radar to follow supported changes and choose your alerts.
Radar
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not publishedBad Theory Labs BTL-4 model card
- Context window
- 262KBad Theory Labs BTL-4 model card
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- Apache-2.0 open-weight checkpoint for local and self-hosted deployment.
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing recordBad Theory Labs BTL-4 model card
- Self-host
- Open weights available; hardware estimate not sourced
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track BTL-4, but the public leaderboard excludes this profile until enough non-generated benchmark coverage is available. Only published rows appear above.
BTL-4 is a open weight model with a 262K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Apache-2.0 open-weight checkpoint for local and self-hosted deployment.
Official exact-value snapshot from Bad Theory Labs' BTL-4 Hugging Face model card. The model is fine-tuned from Ornith-1.0-35B for tool use, software engineering, and long-horizon agent work. BenchLM maps the reported SWE-bench Verified, LiveCodeBench v6, and BFCL v4 AST results; BFCL and LiveCodeBench are in-house runs using official scorers and full splits, while SWE-bench is reported with the official harness.
BTL-4 sits in the BTL family with BTL-3, BTL-4 Compact. 3 of 483 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Last updated September 24, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.
Estimate VRAM from known parametersQuestions
How does BTL-4 perform overall in AI benchmarks?
BTL-4 has 3 source-displayable benchmark rows, but it does not qualify for a public overall rank. The available rows remain visible by category without being converted into a site-wide score. Missing evidence stays blank instead of being estimated from an earlier model.
Is BTL-4 good for coding and programming?
BTL-4 has source-displayable benchmark coverage for coding and programming, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is BTL-4 good for agentic tool use and computer tasks?
BTL-4 has source-displayable benchmark coverage for agentic tool use and computer tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is BTL-4 open source?
BTL-4 is an open-weight model from Bad Theory Labs. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Which sibling models are related to BTL-4?
BTL-4 belongs to the BTL family. Related tracked variants include BTL-3, BTL-4 Compact. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does BTL-4 have full benchmark coverage on BenchLM?
No. BTL-4 currently has 3 source-displayable rows across 483 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of BTL-4?
BTL-4 has a documented context window of 262K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.