Skip to main content
Radar

Provider changes are easy to miss. Radar watches releases, pricing, deprecations, and incidents at the source.Provider changes are easy to miss.

See Radar
BenchLM
Advertise

Model comparison

1-bit Bonsai 4B vs Qwen3.5 397B

Updated August 11, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

1-bit Bonsai 4B

Prism ML

—

Evidence status unavailable

90% interval unavailable

Qwen3.5 397B

Alibaba

56.6/100

Estimated · Public rank #80

90% interval 45.1–68.1

The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. Use the documented cost, context, and runtime rows instead.

0 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

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

    Qwen3.5 397B

    Qwen3.5 397B 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

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    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

    The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. 1-bit Bonsai 4B does not fit this workload in one request. Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate. 1-bit Bonsai 4B has no comparable published API token rate.

    Confidence: rate-fallback

  • Repository review cost

    50K fresh input + 3K output tokens

    Not enough matched evidence

    The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. 1-bit Bonsai 4B does not fit this workload in one request. 1-bit Bonsai 4B has no comparable published API token rate.

    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.

Shared results
0
1-bit Bonsai 4B only
0
Qwen3.5 397B only
38
Like-for-like categories
0 / 8

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Agentic

Not comparable
1-bit Bonsai 4B
Not measured
Qwen3.5 397B
56.5
Weighted basis
0 vs 2 rows
Reading
Not comparable

Coding

Not comparable
1-bit Bonsai 4B
Not measured
Qwen3.5 397B
66.5
Weighted basis
0 vs 2 rows
Reading
Not comparable

Reasoning

Not comparable
1-bit Bonsai 4B
Not measured
Qwen3.5 397B
63.2
Weighted basis
0 vs 1 rows
Reading
Not comparable

Knowledge

Not comparable
1-bit Bonsai 4B
Not measured
Qwen3.5 397B
56.6
Weighted basis
0 vs 4 rows
Reading
Not comparable

Math

Not comparable
1-bit Bonsai 4B
Not measured
Qwen3.5 397B
90.6
Weighted basis
0 vs 2 rows
Reading
Not comparable

Multilingual

Not comparable
1-bit Bonsai 4B
Not measured
Qwen3.5 397B
84.7
Weighted basis
0 vs 1 rows
Reading
Not comparable

Multimodal

Not comparable
1-bit Bonsai 4B
Not measured
Qwen3.5 397B
79.6
Weighted basis
0 vs 2 rows
Reading
Not comparable

Instruction following

Not comparable
1-bit Bonsai 4B
Not measured
Qwen3.5 397B
92.6
Weighted basis
0 vs 1 rows
Reading
Not comparable

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.

A shared-evidence shape is not available.

BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.

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

1-bit Bonsai 4B
Self-hosted; infrastructure cost varies
Fits in one request
Qwen3.5 397B
$0.0024
Fits in one request

1-bit Bonsai 4B has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

1-bit Bonsai 4B
Self-hosted; infrastructure cost varies
Does not fit in one request
Qwen3.5 397B
$0.0408
Fits in one request

1-bit Bonsai 4B does not fit this workload in one request. 1-bit Bonsai 4B has no comparable published API token rate.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

1-bit Bonsai 4B
Self-hosted; infrastructure cost varies
Does not fit in one request
Cached-input rate unavailable
Qwen3.5 397B
$0.168
Does not fit in one request
Cached input priced at the published list-input rate

1-bit Bonsai 4B does not fit this workload in one request. Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate. 1-bit Bonsai 4B 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.

Context window

Maximum documented context; output-token limits may be lower.

1-bit Bonsai 4B

32K

Qwen3.5 397B

128K

API model ID

1-bit Bonsai 4B

Not sourced

Qwen3.5 397B

Not sourced

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

1-bit Bonsai 4B

No comparable hosted API rate

Qwen3.5 397B

Not published

Documented inputs

1-bit Bonsai 4B

Not sourced

Qwen3.5 397B

Not sourced

Documented outputs

1-bit Bonsai 4B

Not sourced

Qwen3.5 397B

Not sourced

Provider availability

1-bit Bonsai 4B

Not sourced

Qwen3.5 397B

Not sourced

Reasoning profile

1-bit Bonsai 4B

Non-Reasoning

Qwen3.5 397B

Non-Reasoning

Weight access

1-bit Bonsai 4B

Open Weight

Qwen3.5 397B

Open Weight

License

1-bit Bonsai 4B

Open Weight

Qwen3.5 397B

Open Weight

Release date

1-bit Bonsai 4B

2026-03-31

Qwen3.5 397B

2026-02-16

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
The public evidence has no benchmark result shared by both models, so it does not support a quality verdict.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
Qwen3.5 397B has the larger documented window (128K).

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 evidence38 rows

Agentic

  • Terminal-Bench 2.0

    1-bit Bonsai 4B—
    Qwen3.5 397B52.5%
    Source

    Not directly comparable

  • BrowseComp

    1-bit Bonsai 4B—
    Qwen3.5 397B62%
    Source

    Not directly comparable

  • Claw-Eval

    1-bit Bonsai 4B—
    Qwen3.5 397B56.8%
    Source

    Not directly comparable

  • QwenClawBench

    1-bit Bonsai 4B—
    Qwen3.5 397B51.8%
    Source

    Not directly comparable

  • τ³-bench results

    1-bit Bonsai 4B—
    Qwen3.5 397B68.4%
    Source

    Not directly comparable

  • VITA-Bench

    1-bit Bonsai 4B—
    Qwen3.5 397B43.7%
    Source

    Not directly comparable

  • DeepPlanning

    1-bit Bonsai 4B—
    Qwen3.5 397B37.6%
    Source

    Not directly comparable

  • Toolathlon

    1-bit Bonsai 4B—
    Qwen3.5 397B36.3%
    Source

    Not directly comparable

  • MCP Atlas

    1-bit Bonsai 4B—
    Qwen3.5 397B46.1%
    Source

    Not directly comparable

  • MCP-Tasks

    1-bit Bonsai 4B—
    Qwen3.5 397B74.2%
    Source

    Not directly comparable

  • WideResearch

    1-bit Bonsai 4B—
    Qwen3.5 397B74.0%
    Source

    Not directly comparable

  • Gert Labs

    1-bit Bonsai 4B—
    Qwen3.5 397B46.76%
    Source

    Not directly comparable

  • ResearchClawBench

    1-bit Bonsai 4B—
    Qwen3.5 397B14.2%
    Source

    Not directly comparable

Coding

  • SWE-bench Verified

    1-bit Bonsai 4B—
    Qwen3.5 397B76.2%
    Source

    Not directly comparable

  • LiveCodeBench v6

    1-bit Bonsai 4B—
    Qwen3.5 397B83.6%
    Source

    Not directly comparable

  • SWE-bench Pro

    1-bit Bonsai 4B—
    Qwen3.5 397B50.9%
    Source

    Not directly comparable

Reasoning

  • LongBench v2

    1-bit Bonsai 4B—
    Qwen3.5 397B63.2%
    Source

    Not directly comparable

  • AI-Needle

    1-bit Bonsai 4B—
    Qwen3.5 397B68.7%
    Source

    Not directly comparable

Knowledge

  • GPQA

    1-bit Bonsai 4B—
    Qwen3.5 397B88.4%
    Source

    Not directly comparable

  • SuperGPQA

    1-bit Bonsai 4B—
    Qwen3.5 397B70.4%
    Source

    Not directly comparable

  • MMLU-Pro

    1-bit Bonsai 4B—
    Qwen3.5 397B87.8%
    Source

    Not directly comparable

  • MMLU-Redux

    1-bit Bonsai 4B—
    Qwen3.5 397B94.9%
    Source

    Not directly comparable

  • C-Eval

    1-bit Bonsai 4B—
    Qwen3.5 397B93%
    Source

    Not directly comparable

  • HLE

    1-bit Bonsai 4B—
    Qwen3.5 397B28.7%
    Source

    Not directly comparable

Math

  • AIME26

    1-bit Bonsai 4B—
    Qwen3.5 397B93.3%
    Source

    Not directly comparable

  • HMMT Feb 2025

    1-bit Bonsai 4B—
    Qwen3.5 397B94.8%
    Source

    Not directly comparable

  • HMMT Nov 2025

    1-bit Bonsai 4B—
    Qwen3.5 397B92.7%
    Source

    Not directly comparable

  • HMMT Feb 2026

    1-bit Bonsai 4B—
    Qwen3.5 397B87.9%
    Source

    Not directly comparable

  • MMAnswerBench

    1-bit Bonsai 4B—
    Qwen3.5 397B80.9%
    Source

    Not directly comparable

Multilingual

  • MMLU-ProX

    1-bit Bonsai 4B—
    Qwen3.5 397B84.7%
    Source

    Not directly comparable

  • NOVA-63

    1-bit Bonsai 4B—
    Qwen3.5 397B59.1%
    Source

    Not directly comparable

Multimodal

  • MMMU-Pro

    1-bit Bonsai 4B—
    Qwen3.5 397B79%
    Source

    Not directly comparable

  • MathVision

    1-bit Bonsai 4B—
    Qwen3.5 397B88.6%
    Source

    Not directly comparable

  • CharXiv

    1-bit Bonsai 4B—
    Qwen3.5 397B80.8%
    Source

    Not directly comparable

  • VideoMMMU

    1-bit Bonsai 4B—
    Qwen3.5 397B84.7%
    Source

    Not directly comparable

  • ScreenSpot Pro

    1-bit Bonsai 4B—
    Qwen3.5 397B65.6%
    Source

    Not directly comparable

  • V*

    1-bit Bonsai 4B—
    Qwen3.5 397B95.8%
    Source

    Not directly comparable

Instruction following

  • IFEval

    1-bit Bonsai 4B—
    Qwen3.5 397B92.6%
    Source

    Not directly comparable

Frequently asked questions

Which is better, 1-bit Bonsai 4B or Qwen3.5 397B?

The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, 1-bit Bonsai 4B or Qwen3.5 397B?

The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.

Which is better for agentic tasks, 1-bit Bonsai 4B or Qwen3.5 397B?

The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.

Which costs less, 1-bit Bonsai 4B or Qwen3.5 397B?

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, 1-bit Bonsai 4B or Qwen3.5 397B?

Qwen3.5 397B has the larger documented context window: 128K, compared with 32K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 11, 2026

Watch 1-bit Bonsai 4B vs Qwen3.5 397B

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.