pi-fireworks-provider 1.1.1 → 1.2.0

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.
Files changed (3) hide show
  1. package/README.md +14 -10
  2. package/package.json +11 -12
  3. package/patch.json +0 -30
package/README.md CHANGED
@@ -15,7 +15,7 @@ _Kimi, MiniMax, GLM, DeepSeek, GPT-OSS — via Fireworks AI's Anthropic Messages
15
15
 
16
16
  ## Features
17
17
 
18
- - **35+ AI Models** including Kimi K2.5, MiniMax M2.5, GLM 4.5/4.7/5, DeepSeek V3.1/V3.2, DeepSeek V4 Flash, and GPT-OSS
18
+ - **39+ AI Models** including Kimi K2.5, MiniMax M2.5, GLM 4.5/4.7/5, DeepSeek V3.1/V3.2, DeepSeek V4 Flash, and GPT-OSS
19
19
  - **Dual API support** via Fireworks AI's Anthropic Messages and OpenAI-compatible completions endpoints (per-model routing, matching pi core's Fireworks provider)
20
20
  - **Service tiers** — toggle Fireworks `priority` vs `standard` per request on supported models (with priority pricing reflected in cost tracking), via a keybinding, `/fireworks-tier`, and a footer status area
21
21
  - **Preserved thinking** — toggle Fireworks' `reasoning_history: "preserved"` so prior assistant reasoning is retained across turns (better multi-turn recall; uses more tokens), via the `/fireworks-settings` panel, with a model-select notification. Matches neuralwatt/makora's settings-only UX, adapted to Fireworks' single global `reasoning_history` knob
@@ -73,8 +73,8 @@ pi
73
73
  |-------|------|---------|------------|------------|-------------|
74
74
  | DeepSeek V3.1 | Text | 164K | 164K | $0.56 | $1.68 |
75
75
  | DeepSeek V3.2 | Text | 164K | 160K | $0.56 | $1.68 |
76
- | DeepSeek V4 Flash | Text | 1.0M | 1.0M | $0.14 | $0.28 |
77
- | DeepSeek V4 Pro | Text | 1.0M | 1.0M | Free | Free |
76
+ | DeepSeek V4 Flash | Text | 1.0M | 384K | $0.14 | $0.28 |
77
+ | DeepSeek V4 Pro | Text | 1.0M | 384K | $1.74 | $3.48 |
78
78
  | DeepSeek V4 Pro (router) | Text | 1.0M | 1.0M | $1.74 | $3.48 |
79
79
  | Gemma 4 26B A4B IT | Text + Image | 262K | 0 | Free | Free |
80
80
  | Gemma 4 31B IT | Text + Image | 262K | 0 | Free | Free |
@@ -82,27 +82,31 @@ pi
82
82
  | GLM 4.5 Air | Text | 131K | 131K | $0.22 | $0.88 |
83
83
  | GLM 4.7 | Text | 203K | 198K | $0.60 | $2.20 |
84
84
  | GLM 5 | Text | 203K | 131K | $1.00 | $3.20 |
85
- | GLM 5 Fast (router) | Text | 203K | 131K | $1.00 | $3.20 |
85
+ | GLM 5 Fast | Text | 203K | 131K | $1.00 | $3.20 |
86
86
  | GLM 5.1 | Text | 203K | 131K | $1.40 | $4.40 |
87
- | GLM 5.1 Fast (router) | Text | 203K | 131K | $1.40 | $4.40 |
88
- | GLM 5.2 | Text | 1.0M | 0 | Free | Free |
87
+ | GLM 5.1 Fast | Text | 203K | 131K | $2.80 | $8.80 |
88
+ | GLM 5.2 | Text | 1.0M | 131K | $1.40 | $4.40 |
89
+ | GLM 5.2 Fast | Text | 1.0M | 131K | $2.10 | $6.60 |
89
90
  | GPT OSS 120B | Text | 131K | 33K | $0.15 | $0.60 |
90
- | GPT OSS 20B | Text | 131K | 33K | $0.05 | $0.20 |
91
+ | GPT OSS 20B | Text | 131K | 33K | $0.07 | $0.30 |
91
92
  | Kimi K2 Instruct | Text | 131K | 16K | $1.00 | $3.00 |
92
93
  | Kimi K2 Thinking | Text | 262K | 256K | $0.60 | $2.50 |
93
94
  | Kimi K2.5 | Text + Image | 262K | 256K | $0.60 | $3.00 |
94
95
  | Kimi K2.5 Fast (router) | Text + Image | 262K | 256K | $0.60 | $3.00 |
95
96
  | Kimi K2.6 | Text + Image | 262K | 262K | $0.95 | $4.00 |
96
97
  | Kimi K2.6 (router) | Text + Image | 262K | 262K | $0.95 | $4.00 |
97
- | Kimi K2.6 Turbo (router) | Text + Image | 262K | 262K | $0.95 | $4.00 |
98
- | Kimi K2.7 Code | Text + Image | 262K | 0 | Free | Free |
98
+ | Kimi K2.6 Fast | Text + Image | 262K | 262K | $2.00 | $8.00 |
99
+ | Kimi K2.6 Turbo | Text + Image | 262K | 262K | $2.00 | $8.00 |
100
+ | Kimi K2.7 Code | Text + Image | 262K | 262K | $0.95 | $4.00 |
101
+ | Kimi K2.7 Code Fast | Text + Image | 262K | 262K | $1.90 | $8.00 |
99
102
  | Llama 3.3 70B Instruct | Text | 131K | 0 | Free | Free |
100
103
  | MiniMax M2.7 (router) | Text | 204K | 0 | $0.30 | $1.20 |
101
- | Minimax M3 | Text + Image | 512K | 0 | Free | Free |
102
104
  | MiniMax-M2.1 | Text | 197K | 200K | $0.30 | $1.20 |
103
105
  | MiniMax-M2.5 | Text | 197K | 197K | $0.30 | $1.20 |
104
106
  | MiniMax-M2.7 | Text | 197K | 197K | $0.30 | $1.20 |
107
+ | MiniMax-M3 | Text | 512K | 512K | $0.30 | $1.20 |
105
108
  | NVIDIA Nemotron 3 Ultra NVFP4 | Text | 262K | 0 | Free | Free |
109
+ | Qwen 3.7 Plus | Text + Image | 262K | 66K | $0.40 | $1.60 |
106
110
  | Qwen3 8B | Text | 41K | 0 | Free | Free |
107
111
  | Qwen3 VL 30B A3B Instruct | Text + Image | 262K | 0 | Free | Free |
108
112
  | Qwen3 VL 30B A3B Thinking | Text + Image | 262K | 0 | Free | Free |
package/package.json CHANGED
@@ -1,18 +1,9 @@
1
1
  {
2
2
  "name": "pi-fireworks-provider",
3
- "version": "1.1.1",
3
+ "version": "1.2.0",
4
4
  "description": "Fireworks AI provider extension for pi - Access Kimi, MiniMax, GLM, DeepSeek, and GPT-OSS models through the Fireworks AI API",
5
5
  "type": "module",
6
6
  "main": "index.ts",
7
- "scripts": {
8
- "clean": "echo 'nothing to clean'",
9
- "build": "echo 'nothing to build'",
10
- "check": "tsc --noEmit && vitest run && knip --no-gitignore",
11
- "lint:dead": "knip --no-gitignore",
12
- "test": "vitest run",
13
- "test:watch": "vitest",
14
- "update-models": "node scripts/update-models.js"
15
- },
16
7
  "keywords": [
17
8
  "pi",
18
9
  "extension",
@@ -47,5 +38,13 @@
47
38
  "./index.ts"
48
39
  ]
49
40
  },
50
- "packageManager": "pnpm@11.6.0"
51
- }
41
+ "scripts": {
42
+ "clean": "echo 'nothing to clean'",
43
+ "build": "echo 'nothing to build'",
44
+ "check": "tsc --noEmit && vitest run && knip --no-gitignore",
45
+ "lint:dead": "knip --no-gitignore",
46
+ "test": "vitest run",
47
+ "test:watch": "vitest",
48
+ "update-models": "node scripts/update-models.js"
49
+ }
50
+ }
package/patch.json CHANGED
@@ -417,22 +417,6 @@
417
417
  "supportsLongCacheRetention": false
418
418
  }
419
419
  },
420
- "accounts/fireworks/models/qwen3p6-plus": {
421
- "name": "Qwen 3.6 Plus",
422
- "reasoning": true,
423
- "input": ["text", "image"],
424
- "cost": {
425
- "input": 0.5,
426
- "output": 3,
427
- "cacheRead": 0.1,
428
- "cacheWrite": 0
429
- },
430
- "contextWindow": 262144,
431
- "maxTokens": 8192,
432
- "compat": {
433
- "supportsReasoningEffort": true
434
- }
435
- },
436
420
  "accounts/fireworks/models/qwen3p7-plus": {
437
421
  "name": "Qwen 3.7 Plus",
438
422
  "api": "anthropic-messages",
@@ -544,20 +528,6 @@
544
528
  "supportsReasoningEffort": true
545
529
  }
546
530
  },
547
- "accounts/fireworks/routers/kimi-k2p5-turbo": {
548
- "name": "Kimi K2.5 Turbo (router)",
549
- "reasoning": true,
550
- "cost": {
551
- "input": 0.6,
552
- "output": 3,
553
- "cacheRead": 0.1,
554
- "cacheWrite": 0
555
- },
556
- "maxTokens": 256000,
557
- "compat": {
558
- "supportsReasoningEffort": true
559
- }
560
- },
561
531
  "accounts/fireworks/routers/kimi-k2p6": {
562
532
  "name": "Kimi K2.6 (router)",
563
533
  "reasoning": true,