Respan
Release: 2026-10-17
Tested on: 2026-10-17 10:59
respan/span-02::none
26B total (4B active)MoEOpen source Summary
Span-02 scores 2.8 on AI BENCHY and ranks #419. It has 10.0 reliability, a 26.1% pass rate, $0.001 total cost, and 1.36s average response time.
What makes Span-02 unique: It stands out most in Anti-AI Tricks, where it ranks #1, while Puzzle Solving is its weakest area at #10.
Model facts
Researched on 2026-10-17
- Parameters
- 26B total (4B active)
- Architecture
- MoE
- Availability
- Open source
- License
- Apache-2.0
The official card identifies a Gemma 4 26B-A4B fine-tune with about 4B active parameters, Apache-2.0 weights and the included Rust/CUDA inference engine. It handles text/JSON decision states as well as optional images; it is not vision-specific. The vendor native API exposes anytime reasoning deadlines, but this OpenRouter checkpoint declares no reasoning controls, so benchmark provider default once. Respan publishes training=false with prompt retention. Fallback date records this onboarding request date.
2.8
Consistency
5.2
10.0
$0.001
Total Output Tokens
0
Total Input Tokens
70,626
Input Price
$0.010 / 1M
Output Price
$0.000 / 1M
Cache Read Price
N/A
Cache Write Price
N/A
Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.
Wrong Tests: 6
Attempt pass rate: 26.1%
Benchmark coverage: 36/69 attempts. Choices supplied by adapters. Supports 12/23 tests. Other tests are unsupported, not failed. Score uses the full suite.
Flaky tests
0
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Charts
Choose the first model, then click a second model to open a side-by-side page.
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
Span-02defaultvsCommand A+lowSpan-02defaultvsd1-ultradefaultSpan-02defaultvsCommand A+noneSpan-02defaultvsJev 1.13defaultSpan-02defaultvsCommand A+highSpan-02defaultvsDecider V1.2defaultSpan-02defaultvsXing4.0 29b A4bnoneFree AvailableSpan-02defaultvsDecider V1.1 27BdefaultSpan-02defaultvsCommand A+mediumSpan-02defaultvsDrex v1.5default
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Agentic | 0.0 | 0.0 | |
| Anti-AI Tricks | 5.8 | 7.5 | |
| Coding | 4.9 | 6.7 | |
| Combined | 0.0 | 0.0 | |
| Data parsing and extraction | 6.5 | 10.0 | |
| Domain specific | 5.3 | 10.0 | |
| General Intelligence | 0.0 | 0.0 | |
| Instructions following | 5.0 | 5.0 | |
| Puzzle Solving | 1.0 | 3.3 | |
| Tool Calling | 0.0 | 0.0 | |
| Trivia | 0.0 | 0.0 |