Model comparison
Claude Haiku 4.5 vs gpt-oss-120b
ClaudeCompare published specifications, vendor API list prices and measured Borg performance. Both models are included in Borg.
Claude Haiku 4.5
gpt-oss-120b
Included in Borg
At a glance
| Specification | Claude Haiku 4.5 | gpt-oss-120b |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Release | 2025-10-15 Vendor source | 2025-08-05 Vendor source |
| Context | 200,000 tokens Vendor source | 131,072 tokens Vendor source |
| Modalities | Input: text, image; output: text | Input: text; output: text |
| Reasoning / thinking modes | Yes / Not published | Yes / Not published |
| Open weights | No | Yes |
| Vendor API list price / 1M tokens | $1 input / $5 output | — |
| Borg | Included in Borg subscription | Included in Borg subscription |
Anthropic
OpenAINot published by the vendor: vendor api list price / 1m tokens.
Measured on Borg
Metrics are not directly comparable to vendor benchmark claims. Models served different requests; this is observed gateway traffic, not a controlled head-to-head test.
Time to first token (gateway proxy)
Output tokens per second (whole turn)
| Metric | Claude Haiku 4.5 | gpt-oss-120b |
|---|---|---|
| Time to first token (gateway proxy) p50 | 889 ms (6197 samples) | 6121 ms (15830 samples) |
| Tokens per second (end to end) p50 | 64.9 output tokens per second (226302 samples) | 46.8 output tokens per second (92214 samples) |
| Request success rate | Not published | 100% (21212 samples) |
| Coding bench pass rate | Not published | Not published |
Measurement window, samples and methodology
The fleet reference is the median of published model medians, not a request-weighted result. Prompt lengths, reasoning settings and traffic differ between models.
Claude Haiku 4.5: Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 289125 requests. Minimum 100 samples per published request metric.
gpt-oss-120b: Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 93142 requests. Minimum 100 samples per published request metric.
Completion speed covers the whole turn including queueing and reasoning time; it is not a streaming tokens-per-second figure.
- Time to first token (gateway proxy): time until the first streamed response event reaches the gateway; our closest available measure of time to first token
- Tokens per second (end to end): tokens per second, end to end: provider-reported output tokens divided by the whole turn's elapsed time, including queueing and reasoning; not a streaming per-token rate
How to choose
Claude Haiku 4.5 is listed for high-volume classification and tagging, whereas gpt-oss-120b is listed for private self-hosted reasoning. Choosing between Claude Haiku 4.5 and gpt-oss-120b requires checking published specifications against observed Borg traffic for your workflow.
Key differences
| Spec | Claude Haiku 4.5 | gpt-oss-120b |
|---|---|---|
| Context window | 200,000 tokens Vendor source Larger context | 131,072 tokens Vendor source |
| Reasoning modes | Yes / Not published | Yes / Not published |
| Release date | 2025-10-15 Vendor source | 2025-08-05 Vendor source |
| API price input / 1M | 1 | Not published |
| API price output / 1M | 5 | Not published |
| Time to first token (gateway proxy) | 889 ms (6197 samples) Lower median | 6121 ms (15830 samples) |
| Output tokens per second (whole turn) | 64.9 output tokens/s (whole turn) (226302 samples) Higher measured rate | 46.8 output tokens/s (whole turn) (92214 samples) |
| Request success | Not published | 100% (21212 samples) |
| Coding bench | Not published | Not published |
FAQ
How do Claude Haiku 4.5 and gpt-oss-120b compare for coding?
Claude Haiku 4.5 has a Borg coding benchmark of not published; gpt-oss-120b has not published. The Claude Haiku 4.5–gpt-oss-120b coding comparison cannot establish an overall winner from this small test alone.
How quickly do Claude Haiku 4.5 and gpt-oss-120b start responding on Borg?
Borg median first response for Claude Haiku 4.5 is 889 ms; for gpt-oss-120b it is 6,121 ms. The Claude Haiku 4.5–gpt-oss-120b gateway medians are observations, not guaranteed latency for individual requests.
What are the API prices for Claude Haiku 4.5 and gpt-oss-120b?
The listed API rate for Claude Haiku 4.5 is $1 input / $5 output per 1M tokens; for gpt-oss-120b it is not published. Neither Claude Haiku 4.5 nor gpt-oss-120b has its Borg subscription charge represented by those provider API rates. Claude Haiku 4.5 vendor source
Gemini