Use short context for GPT-6 Astra
- commit
- 0d096bd2c70f515e33c938ba01a82d63a60bb6a9
- author
- patriceckhart · · 2026-09-05 05:56 UTC
- parent
- 8233420
- signature
- unsigned
- refs
- tag: v0.3.63
4 files · +8 -8
M docs/providers.md +3 -3
| @@ -100,9 +100,9 @@ zot --provider github-copilot --model gpt-6-astra | |||
| 100 | 100 | ``` | |
| 101 | 101 | ||
| 102 | 102 | Astra uses the Responses API and supports `low`, `medium`, `high`, `xhigh`, | |
| 103 | - | and `max` reasoning. Public API and Copilot catalog entries include its | |
| 104 | - | 1,050,000-token context window; the OpenAI Codex subscription entry uses the | |
| 105 | - | backend's 272,000-token default. All entries include the 128,000-token output | |
| 103 | + | and `max` reasoning. OpenAI API-key and Codex subscription entries use the | |
| 104 | + | 272,000-token short-context tier by default; the Copilot entry uses a | |
| 105 | + | 1,050,000-token context window. All entries include the 128,000-token output | |
| 106 | 106 | limit and standard token prices, including the higher rates for prompts above | |
| 107 | 107 | 272,000 tokens. Subscription access depends on your | |
| 108 | 108 | plan, rollout, and organization policy; catalog availability does not grant access. | |
| @@ -100,9 +100,9 @@ zot --provider github-copilot --model gpt-6-astra | |||||
| 100 | ``` | 100 | ``` | ||
| 101 | 101 | ||||
| 102 | Astra uses the Responses API and supports `low`, `medium`, `high`, `xhigh`, | 102 | Astra uses the Responses API and supports `low`, `medium`, `high`, `xhigh`, | ||
| 103 | - | and `max` reasoning. Public API and Copilot catalog entries include its | 103 | + | and `max` reasoning. OpenAI API-key and Codex subscription entries use the |
| 104 | - | 1,050,000-token context window; the OpenAI Codex subscription entry uses the | 104 | + | 272,000-token short-context tier by default; the Copilot entry uses a |
| 105 | - | backend's 272,000-token default. All entries include the 128,000-token output | 105 | + | 1,050,000-token context window. All entries include the 128,000-token output |
| 106 | limit and standard token prices, including the higher rates for prompts above | 106 | limit and standard token prices, including the higher rates for prompts above | ||
| 107 | 272,000 tokens. Subscription access depends on your | 107 | 272,000 tokens. Subscription access depends on your | ||
| 108 | plan, rollout, and organization policy; catalog availability does not grant access. | 108 | plan, rollout, and organization policy; catalog availability does not grant access. | ||
M packages/provider/extra_models.go +1 -1
| @@ -49,7 +49,7 @@ var supplementCatalog = []Model{ | |||
| 49 | 49 | ||
| 50 | 50 | // ----- openai-responses (public OpenAI Responses API) ----- | |
| 51 | 51 | {Provider: "openai-responses", ID: "gpt-6-astra", DisplayName: "GPT-6 Astra (Responses)", API: APIResponses, | |
| 52 | - | ContextWindow: 1050000, MaxOutput: 128000, Reasoning: true, | |
| 52 | + | ContextWindow: 272000, MaxOutput: 128000, Reasoning: true, | |
| 53 | 53 | PriceInput: 10, PriceOutput: 50, PriceCacheRead: 1, PriceCacheWrite: 12.5, | |
| 54 | 54 | PriceTierInputTokens: 272000, | |
| 55 | 55 | PriceInputAbove: 20, PriceOutputAbove: 75, PriceCacheReadAbove: 2, PriceCacheWriteAbove: 25}, | |
| @@ -49,7 +49,7 @@ var supplementCatalog = []Model{ | |||||
| 49 | 49 | ||||
| 50 | // ----- openai-responses (public OpenAI Responses API) ----- | 50 | // ----- openai-responses (public OpenAI Responses API) ----- | ||
| 51 | {Provider: "openai-responses", ID: "gpt-6-astra", DisplayName: "GPT-6 Astra (Responses)", API: APIResponses, | 51 | {Provider: "openai-responses", ID: "gpt-6-astra", DisplayName: "GPT-6 Astra (Responses)", API: APIResponses, | ||
| 52 | - | ContextWindow: 1050000, MaxOutput: 128000, Reasoning: true, | 52 | + | ContextWindow: 272000, MaxOutput: 128000, Reasoning: true, |
| 53 | PriceInput: 10, PriceOutput: 50, PriceCacheRead: 1, PriceCacheWrite: 12.5, | 53 | PriceInput: 10, PriceOutput: 50, PriceCacheRead: 1, PriceCacheWrite: 12.5, | ||
| 54 | PriceTierInputTokens: 272000, | 54 | PriceTierInputTokens: 272000, | ||
| 55 | PriceInputAbove: 20, PriceOutputAbove: 75, PriceCacheReadAbove: 2, PriceCacheWriteAbove: 25}, | 55 | PriceInputAbove: 20, PriceOutputAbove: 75, PriceCacheReadAbove: 2, PriceCacheWriteAbove: 25}, | ||
M packages/provider/gpt6_astra_test.go +3 -3
| @@ -13,9 +13,9 @@ func TestGPT6AstraCatalog(t *testing.T) { | |||
| 13 | 13 | if err != nil { | |
| 14 | 14 | t.Fatal(err) | |
| 15 | 15 | } | |
| 16 | - | wantContext := 1050000 | |
| 17 | - | if name == "openai-codex" { | |
| 18 | - | wantContext = 272000 | |
| 16 | + | wantContext := 272000 | |
| 17 | + | if name == "github-copilot" { | |
| 18 | + | wantContext = 1050000 | |
| 19 | 19 | } | |
| 20 | 20 | if m.API != APIResponses || m.ContextWindow != wantContext || m.MaxOutput != 128000 || !m.Reasoning || m.Speculative { | |
| 21 | 21 | t.Fatalf("unexpected model capabilities: %+v", m) | |
| @@ -13,9 +13,9 @@ func TestGPT6AstraCatalog(t *testing.T) { | |||||
| 13 | if err != nil { | 13 | if err != nil { | ||
| 14 | t.Fatal(err) | 14 | t.Fatal(err) | ||
| 15 | } | 15 | } | ||
| 16 | - | wantContext := 1050000 | 16 | + | wantContext := 272000 |
| 17 | - | if name == "openai-codex" { | 17 | + | if name == "github-copilot" { |
| 18 | - | wantContext = 272000 | 18 | + | wantContext = 1050000 |
| 19 | } | 19 | } | ||
| 20 | if m.API != APIResponses || m.ContextWindow != wantContext || m.MaxOutput != 128000 || !m.Reasoning || m.Speculative { | 20 | if m.API != APIResponses || m.ContextWindow != wantContext || m.MaxOutput != 128000 || !m.Reasoning || m.Speculative { | ||
| 21 | t.Fatalf("unexpected model capabilities: %+v", m) | 21 | t.Fatalf("unexpected model capabilities: %+v", m) | ||
M packages/provider/models.go +1 -1
| @@ -379,7 +379,7 @@ var Catalog = []Model{ | |||
| 379 | 379 | // ---- OpenAI / GPT-6 ---- | |
| 380 | 380 | { | |
| 381 | 381 | Provider: "openai", ID: "gpt-6-astra", DisplayName: "GPT-6 Astra", API: APIResponses, | |
| 382 | - | ContextWindow: 1050000, MaxOutput: 128000, Reasoning: true, | |
| 382 | + | ContextWindow: 272000, MaxOutput: 128000, Reasoning: true, | |
| 383 | 383 | PriceInput: 10, PriceOutput: 50, PriceCacheRead: 1, PriceCacheWrite: 12.5, | |
| 384 | 384 | PriceTierInputTokens: 272000, | |
| 385 | 385 | PriceInputAbove: 20, PriceOutputAbove: 75, PriceCacheReadAbove: 2, PriceCacheWriteAbove: 25, | |
| @@ -379,7 +379,7 @@ var Catalog = []Model{ | |||||
| 379 | // ---- OpenAI / GPT-6 ---- | 379 | // ---- OpenAI / GPT-6 ---- | ||
| 380 | { | 380 | { | ||
| 381 | Provider: "openai", ID: "gpt-6-astra", DisplayName: "GPT-6 Astra", API: APIResponses, | 381 | Provider: "openai", ID: "gpt-6-astra", DisplayName: "GPT-6 Astra", API: APIResponses, | ||
| 382 | - | ContextWindow: 1050000, MaxOutput: 128000, Reasoning: true, | 382 | + | ContextWindow: 272000, MaxOutput: 128000, Reasoning: true, |
| 383 | PriceInput: 10, PriceOutput: 50, PriceCacheRead: 1, PriceCacheWrite: 12.5, | 383 | PriceInput: 10, PriceOutput: 50, PriceCacheRead: 1, PriceCacheWrite: 12.5, | ||
| 384 | PriceTierInputTokens: 272000, | 384 | PriceTierInputTokens: 272000, | ||
| 385 | PriceInputAbove: 20, PriceOutputAbove: 75, PriceCacheReadAbove: 2, PriceCacheWriteAbove: 25, | 385 | PriceInputAbove: 20, PriceOutputAbove: 75, PriceCacheReadAbove: 2, PriceCacheWriteAbove: 25, | ||