@diegopetrucci/pi-fast 0.1.4 → 0.1.6
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +36 -3
- package/index.ts +6 -2
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -38,17 +38,46 @@ Direct OpenAI API Fast mode requires all of the following:
|
|
|
38
38
|
|
|
39
39
|
- Provider `openai`.
|
|
40
40
|
- API `openai-responses` or `openai-completions`.
|
|
41
|
-
- Model `gpt-6-astra`, `gpt-6-sol`, or `gpt-6-luna`.
|
|
41
|
+
- Model `gpt-6.1-sol`, `gpt-6-astra`, `gpt-6-sol`, or `gpt-6-luna`.
|
|
42
|
+
- API-key auth or Pi's `openai` ChatGPT OAuth (`Sign in with ChatGPT`).
|
|
42
43
|
- No existing `service_tier` field in the request payload.
|
|
43
44
|
|
|
45
|
+
The direct OpenAI route stays on OpenAI's direct API contract for both API-key
|
|
46
|
+
auth and Pi's `openai` ChatGPT OAuth. Acceptance of `service_tier: "fast"` by
|
|
47
|
+
Pi's `openai` ChatGPT OAuth sign-in endpoint has not been live-verified and
|
|
48
|
+
remains deferred to manual gate `pel-71t6`. This direct route is never forwarded
|
|
49
|
+
to legacy `openai-codex` or its subscription-usage endpoint. The documented
|
|
50
|
+
request shape is support evidence, not live certification; this change was
|
|
51
|
+
validated only with mocked eligibility and request payloads, not live or paid
|
|
52
|
+
provider calls.
|
|
53
|
+
|
|
44
54
|
OpenAI Codex Fast mode requires all of the following:
|
|
45
55
|
|
|
46
56
|
- Provider `openai-codex`.
|
|
47
57
|
- API `openai-codex-responses`.
|
|
48
|
-
- Model `gpt-5.5`, `gpt-5.6-sol`, `gpt-5.6-terra`,
|
|
58
|
+
- Model `gpt-5.5`, `gpt-5.6-sol`, `gpt-5.6-terra`, `gpt-5.6-luna`, `gpt-6-astra`, `gpt-6.1-sol`, `gpt-6-sol`, or `gpt-6-luna`.
|
|
49
59
|
- ChatGPT OAuth/subscription auth, not API-key auth.
|
|
50
60
|
- No existing `service_tier` field in the request payload.
|
|
51
61
|
|
|
62
|
+
Codex evidence, 2026-10-03: the [maintainer's PR #103 review](https://github.com/diegopetrucci/pi-extensions/pull/103)
|
|
63
|
+
reports that the Codex model list advertises Fast as `priority` for `gpt-6-astra`,
|
|
64
|
+
`gpt-6.1-sol`, `gpt-6-sol`, and `gpt-6-luna`. The official Codex CLI
|
|
65
|
+
[sends Fast as `priority`](https://github.com/openai/codex/blob/b741e480e203f037ca726bc2a76d99a8e8668e66/codex-rs/protocol/src/config_types.rs#L539-L553).
|
|
66
|
+
The maintainer's live check through Pi 1.0.0 on that date returned HTTP 200 and
|
|
67
|
+
completed normally for `gpt-6-astra` and `gpt-6.1-sol`; `gpt-6-sol` and
|
|
68
|
+
`gpt-6-luna` are enabled based on the model list only. In that check,
|
|
69
|
+
`response.completed` always reported `service_tier: "default"`, with or without
|
|
70
|
+
`priority`, so the response cannot confirm Fast was applied.
|
|
71
|
+
|
|
72
|
+
Pi trusts the requested tier and applies a 2× multiplier to its displayed cost
|
|
73
|
+
estimate for `priority`; this does not measure included subscription allowance
|
|
74
|
+
consumption or confirm Fast was applied. [OpenAI's Codex speed documentation](https://developers.openai.com/codex/speed),
|
|
75
|
+
checked on 2026-10-03, lists `gpt-6-sol` and `gpt-6-luna` among supported Fast
|
|
76
|
+
models where available. It specifies 2.5× the Standard rate for included
|
|
77
|
+
subscription limits and 2× for purchased credits and Enterprise pay-as-you-go
|
|
78
|
+
billing. Treating Pi's 2× estimate as a subscription-limit multiplier would
|
|
79
|
+
understate the documented allowance consumption.
|
|
80
|
+
|
|
52
81
|
Anthropic Fast mode requires all of the following:
|
|
53
82
|
|
|
54
83
|
- Provider `anthropic`.
|
|
@@ -129,7 +158,11 @@ The collection package loads this unified extension instead of `openai-fast` and
|
|
|
129
158
|
## Notes
|
|
130
159
|
|
|
131
160
|
- Anthropic Fast mode has separate rate limits, costs more than standard speed, and does not share prompt-cache prefixes with standard-speed requests.
|
|
132
|
-
- Direct OpenAI Fast mode is limited to the confirmed
|
|
161
|
+
- Direct OpenAI Fast mode is limited to the confirmed GPT-6 API models above; Fast pricing, availability, and rate limits remain provider- and account-dependent.
|
|
162
|
+
- Authoritative support evidence: [OpenAI Fast mode](https://developers.openai.com/api/docs/guides/fast-mode) explicitly uses `gpt-6.1-sol` with `service_tier: "fast"` and documents `priority` as an equivalent alias for supported API models; the [GPT-6.1 Sol model page](https://developers.openai.com/api/docs/models/gpt-6.1-sol) documents the model itself.
|
|
163
|
+
- Codex Fast availability depends on plan, client, workspace settings, and rollout. Local eligibility is not live-provider certification.
|
|
164
|
+
- Pi's provider catalog remains the source of reported model pricing, including any tiered pricing; this extension does not apply guessed discounts or rewrite cost metadata.
|
|
133
165
|
- OpenAI Codex Fast mode intentionally does not affect API-key models.
|
|
166
|
+
- The retired `openai-fast` and `claude-fast` packages remain legacy standalone boundaries; new provider/auth behavior belongs only to this unified extension.
|
|
134
167
|
- Existing `speed` and `service_tier` fields are never overwritten.
|
|
135
168
|
- Pi's model catalog may not include a Fast-mode premium. Cost accounting depends on the provider reporting the effective tier in its streamed response, and this extension does not patch usage totals.
|
package/index.ts
CHANGED
|
@@ -25,11 +25,15 @@ const OPENAI_SUPPORTED_MODELS = new Set([
|
|
|
25
25
|
"gpt-5.6-sol",
|
|
26
26
|
"gpt-5.6-terra",
|
|
27
27
|
"gpt-5.6-luna",
|
|
28
|
+
"gpt-6-astra",
|
|
29
|
+
"gpt-6.1-sol",
|
|
30
|
+
"gpt-6-sol",
|
|
31
|
+
"gpt-6-luna",
|
|
28
32
|
]);
|
|
29
33
|
const OPENAI_API_PROVIDER_ID = "openai";
|
|
30
34
|
const OPENAI_API_IDS = new Set(["openai-responses", "openai-completions"]);
|
|
31
35
|
const OPENAI_API_FAST_SERVICE_TIER = "fast";
|
|
32
|
-
const OPENAI_API_SUPPORTED_MODELS = new Set(["gpt-6-astra", "gpt-6-sol", "gpt-6-luna"]);
|
|
36
|
+
const OPENAI_API_SUPPORTED_MODELS = new Set(["gpt-6.1-sol", "gpt-6-astra", "gpt-6-sol", "gpt-6-luna"]);
|
|
33
37
|
|
|
34
38
|
const DEFAULT_CONFIG: FastConfig = {
|
|
35
39
|
enabled: false,
|
|
@@ -216,7 +220,7 @@ function getEligibility(ctx: ExtensionContext): Eligibility {
|
|
|
216
220
|
return {
|
|
217
221
|
eligible: false,
|
|
218
222
|
modelKey: key,
|
|
219
|
-
reason: "Fast mode is only enabled for GPT-6 Astra, GPT-6 Sol, and GPT-6 Luna on direct OpenAI APIs",
|
|
223
|
+
reason: "Fast mode is only enabled for GPT-6.1 Sol, GPT-6 Astra, GPT-6 Sol, and GPT-6 Luna on direct OpenAI APIs",
|
|
220
224
|
};
|
|
221
225
|
}
|
|
222
226
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@diegopetrucci/pi-fast",
|
|
3
|
-
"version": "0.1.
|
|
3
|
+
"version": "0.1.6",
|
|
4
4
|
"description": "A pi extension that enables Fast mode for supported direct OpenAI API, OpenAI Codex, and Anthropic Claude models based on the currently selected model.",
|
|
5
5
|
"keywords": [
|
|
6
6
|
"pi-package",
|