@blockrun/clawrouter 0.9.39 → 0.10.0

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 (7) hide show
  1. package/README.md +18 -13
  2. package/dist/cli.js +175 -68
  3. package/dist/cli.js.map +1 -1
  4. package/dist/index.d.ts +2 -0
  5. package/dist/index.js +175 -68
  6. package/dist/index.js.map +1 -1
  7. package/package.json +2 -2
package/README.md CHANGED
@@ -6,11 +6,11 @@
6
6
 
7
7
  Route every request to the right model at the right price.<br>
8
8
  15-dimension scoring, <1ms local routing, optimized for autonomous agents.<br>
9
- One wallet, 30+ models, zero API keys.
9
+ One wallet, 38+ models, zero API keys.
10
10
 
11
11
  <img src="https://img.shields.io/badge/🚀_92%25_Cost_Savings-success?style=for-the-badge" alt="92% savings">&nbsp;
12
12
  <img src="https://img.shields.io/badge/🔑_Zero_API_Keys-blue?style=for-the-badge" alt="No API keys">&nbsp;
13
- <img src="https://img.shields.io/badge/🤖_30+_Models-purple?style=for-the-badge" alt="30+ models">&nbsp;
13
+ <img src="https://img.shields.io/badge/🤖_38+_Models-purple?style=for-the-badge" alt="38+ models">&nbsp;
14
14
  <img src="https://img.shields.io/badge/💰_Non--Custodial-orange?style=for-the-badge" alt="Non-custodial">&nbsp;
15
15
  <img src="https://img.shields.io/badge/⚡_<1ms_Routing-yellow?style=for-the-badge" alt="Fast routing">
16
16
 
@@ -85,12 +85,12 @@ Choose your routing strategy with `/model <profile>`:
85
85
  Request → Weighted Scorer (15 dimensions) → Tier → Cheapest Model → Response
86
86
  ```
87
87
 
88
- | Tier | ECO Model | AUTO Model | PREMIUM Model |
89
- | --------- | ------------------------------ | ---------------------------- | ------------------------- |
90
- | SIMPLE | nvidia/gpt-oss-120b (FREE) | kimi-k2.5 ($0.50/$2.40) | kimi-k2.5 |
91
- | MEDIUM | gemini-2.5-flash ($0.15/$0.60) | grok-code-fast ($0.20/$1.50) | gpt-5.2-codex ($2.50/$10) |
92
- | COMPLEX | gemini-2.5-flash ($0.15/$0.60) | gemini-3-pro ($2/$12) | claude-opus-4 ($15/$75) |
93
- | REASONING | grok-4-fast ($0.20/$0.50) | grok-4-fast ($0.20/$0.50) | claude-sonnet-4 ($3/$15) |
88
+ | Tier | ECO Model | AUTO Model | PREMIUM Model |
89
+ | --------- | ---------------------------------- | ---------------------------- | ---------------------------- |
90
+ | SIMPLE | nvidia/gpt-oss-120b (FREE) | kimi-k2.5 ($0.60/$3.00) | kimi-k2.5 |
91
+ | MEDIUM | gemini-2.5-flash-lite ($0.10/$0.40) | grok-code-fast ($0.20/$1.50) | gpt-5.2-codex ($1.75/$14.00) |
92
+ | COMPLEX | gemini-2.5-flash-lite ($0.10/$0.40) | gemini-3.1-pro ($2/$12) | claude-opus-4.6 ($5/$25) |
93
+ | REASONING | grok-4-fast ($0.20/$0.50) | grok-4-fast ($0.20/$0.50) | claude-sonnet-4.6 ($3/$15) |
94
94
 
95
95
  **Blended average: $2.05/M** vs $25/M for Claude Opus = **92% savings**
96
96
 
@@ -98,7 +98,7 @@ Request → Weighted Scorer (15 dimensions) → Tier → Cheapest Model → Resp
98
98
 
99
99
  ## 💰 Models & Pricing
100
100
 
101
- 30+ models across 7 providers, one wallet:
101
+ 38+ models across 7 providers, one wallet:
102
102
 
103
103
  <details>
104
104
  <summary><strong>Click to expand full model list</strong></summary>
@@ -110,6 +110,8 @@ Request → Weighted Scorer (15 dimensions) → Tier → Cheapest Model → Resp
110
110
  | gpt-4o | $2.50 | $10.00 | 128K | |
111
111
  | gpt-4o-mini | $0.15 | $0.60 | 128K | |
112
112
  | gpt-oss-120b | **FREE** | **FREE** | 128K | |
113
+ | o1 | $15.00 | $60.00 | 200K | \* |
114
+ | o1-mini | $1.10 | $4.40 | 128K | \* |
113
115
  | o3 | $2.00 | $8.00 | 200K | \* |
114
116
  | o4-mini | $1.10 | $4.40 | 128K | \* |
115
117
  | **Anthropic** | | | | |
@@ -117,18 +119,21 @@ Request → Weighted Scorer (15 dimensions) → Tier → Cheapest Model → Resp
117
119
  | claude-sonnet-4.6 | $3.00 | $15.00 | 200K | \* |
118
120
  | claude-haiku-4.5 | $1.00 | $5.00 | 200K | |
119
121
  | **Google** | | | | |
122
+ | gemini-3.1-pro-preview | $2.00 | $12.00 | 1M | \* |
120
123
  | gemini-3-pro-preview | $2.00 | $12.00 | 1M | \* |
124
+ | gemini-3-flash-preview | $0.50 | $3.00 | 1M | |
121
125
  | gemini-2.5-pro | $1.25 | $10.00 | 1M | \* |
122
- | gemini-2.5-flash | $0.15 | $0.60 | 1M | |
126
+ | gemini-2.5-flash | $0.30 | $2.50 | 1M | |
127
+ | gemini-2.5-flash-lite | $0.10 | $0.40 | 1M | |
123
128
  | **DeepSeek** | | | | |
124
- | deepseek-chat | $0.14 | $0.28 | 128K | |
125
- | deepseek-reasoner | $0.55 | $2.19 | 128K | \* |
129
+ | deepseek-chat | $0.28 | $0.42 | 128K | |
130
+ | deepseek-reasoner | $0.28 | $0.42 | 128K | \* |
126
131
  | **xAI** | | | | |
127
132
  | grok-4-0709 | $0.20 | $1.50 | 131K | \* |
128
133
  | grok-4-1-fast-reasoning | $0.20 | $0.50 | 131K | \* |
129
134
  | grok-code-fast-1 | $0.20 | $1.50 | 131K | |
130
135
  | **Moonshot** | | | | |
131
- | kimi-k2.5 | $0.50 | $2.40 | 262K | \* |
136
+ | kimi-k2.5 | $0.60 | $3.00 | 262K | \* |
132
137
  | **MiniMax** | | | | |
133
138
  | minimax-m2.5 | $0.30 | $1.20 | 205K | \* |
134
139
 
package/dist/cli.js CHANGED
@@ -1199,10 +1199,10 @@ var DEFAULT_ROUTING_CONFIG = {
1199
1199
  tiers: {
1200
1200
  SIMPLE: {
1201
1201
  primary: "moonshot/kimi-k2.5",
1202
- // $0.50/$2.40 - best quality/price for simple tasks
1202
+ // $0.60/$3.00 - best quality/price for simple tasks
1203
1203
  fallback: [
1204
- "google/gemini-2.5-flash",
1205
- // 1M context, cost-effective
1204
+ "google/gemini-2.5-flash-lite",
1205
+ // 1M context, ultra cheap ($0.10/$0.40)
1206
1206
  "nvidia/gpt-oss-120b",
1207
1207
  // FREE fallback
1208
1208
  "deepseek/deepseek-chat"
@@ -1212,26 +1212,28 @@ var DEFAULT_ROUTING_CONFIG = {
1212
1212
  primary: "xai/grok-code-fast-1",
1213
1213
  // Code specialist, $0.20/$1.50
1214
1214
  fallback: [
1215
- "google/gemini-2.5-flash",
1216
- // 1M context, cost-effective
1215
+ "google/gemini-2.5-flash-lite",
1216
+ // 1M context, ultra cheap ($0.10/$0.40)
1217
1217
  "deepseek/deepseek-chat",
1218
1218
  "xai/grok-4-1-fast-non-reasoning"
1219
1219
  // Upgraded Grok 4.1
1220
1220
  ]
1221
1221
  },
1222
1222
  COMPLEX: {
1223
- primary: "google/gemini-3-pro-preview",
1224
- // Latest Gemini - upgraded from 2.5
1223
+ primary: "google/gemini-3.1-pro-preview",
1224
+ // Newest Gemini 3.1 - upgraded from 3.0
1225
1225
  fallback: [
1226
- "google/gemini-2.5-flash",
1227
- // CRITICAL: 1M context, cheap failsafe before expensive models
1226
+ "google/gemini-2.5-flash-lite",
1227
+ // CRITICAL: 1M context, ultra-cheap failsafe ($0.10/$0.40)
1228
+ "google/gemini-3-pro-preview",
1229
+ // 3.0 fallback
1228
1230
  "google/gemini-2.5-pro",
1229
1231
  "deepseek/deepseek-chat",
1230
1232
  "xai/grok-4-0709",
1231
1233
  "openai/gpt-5.2",
1232
1234
  // Newer and cheaper input than gpt-4o
1233
1235
  "openai/gpt-4o",
1234
- "claude-sonnet-4"
1236
+ "anthropic/claude-sonnet-4.6"
1235
1237
  ]
1236
1238
  },
1237
1239
  REASONING: {
@@ -1251,17 +1253,17 @@ var DEFAULT_ROUTING_CONFIG = {
1251
1253
  SIMPLE: {
1252
1254
  primary: "nvidia/gpt-oss-120b",
1253
1255
  // FREE! $0.00/$0.00
1254
- fallback: ["google/gemini-2.5-flash", "deepseek/deepseek-chat"]
1256
+ fallback: ["google/gemini-2.5-flash-lite", "deepseek/deepseek-chat"]
1255
1257
  },
1256
1258
  MEDIUM: {
1257
- primary: "google/gemini-2.5-flash",
1258
- // $0.15/$0.60 - cheapest capable
1259
+ primary: "google/gemini-2.5-flash-lite",
1260
+ // $0.10/$0.40 - cheapest capable with 1M context
1259
1261
  fallback: ["deepseek/deepseek-chat", "nvidia/gpt-oss-120b"]
1260
1262
  },
1261
1263
  COMPLEX: {
1262
- primary: "google/gemini-2.5-flash",
1263
- // $0.15/$0.60 - 1M context handles complexity
1264
- fallback: ["deepseek/deepseek-chat", "xai/grok-4-0709"]
1264
+ primary: "google/gemini-2.5-flash-lite",
1265
+ // $0.10/$0.40 - 1M context handles complexity
1266
+ fallback: ["google/gemini-2.5-flash", "deepseek/deepseek-chat", "xai/grok-4-0709"]
1265
1267
  },
1266
1268
  REASONING: {
1267
1269
  primary: "xai/grok-4-1-fast-reasoning",
@@ -1274,8 +1276,8 @@ var DEFAULT_ROUTING_CONFIG = {
1274
1276
  premiumTiers: {
1275
1277
  SIMPLE: {
1276
1278
  primary: "moonshot/kimi-k2.5",
1277
- // $0.50/$2.40 - good for simple coding
1278
- fallback: ["claude-haiku-4.5", "google/gemini-2.5-flash", "xai/grok-code-fast-1"]
1279
+ // $0.60/$3.00 - good for simple coding
1280
+ fallback: ["anthropic/claude-haiku-4.5", "google/gemini-2.5-flash-lite", "xai/grok-code-fast-1"]
1279
1281
  },
1280
1282
  MEDIUM: {
1281
1283
  primary: "openai/gpt-5.2-codex",
@@ -1284,26 +1286,28 @@ var DEFAULT_ROUTING_CONFIG = {
1284
1286
  "moonshot/kimi-k2.5",
1285
1287
  "google/gemini-2.5-pro",
1286
1288
  "xai/grok-4-0709",
1287
- "claude-sonnet-4"
1289
+ "anthropic/claude-sonnet-4.6"
1288
1290
  ]
1289
1291
  },
1290
1292
  COMPLEX: {
1291
- primary: "claude-opus-4",
1293
+ primary: "anthropic/claude-opus-4.6",
1292
1294
  // Best quality for complex tasks
1293
1295
  fallback: [
1294
1296
  "openai/gpt-5.2-codex",
1295
- "claude-opus-4",
1296
- "claude-sonnet-4",
1297
+ "anthropic/claude-opus-4.6",
1298
+ "anthropic/claude-sonnet-4.6",
1299
+ "google/gemini-3.1-pro-preview",
1300
+ // Newest Gemini
1297
1301
  "google/gemini-3-pro-preview",
1298
1302
  "moonshot/kimi-k2.5"
1299
1303
  ]
1300
1304
  },
1301
1305
  REASONING: {
1302
- primary: "claude-sonnet-4",
1306
+ primary: "anthropic/claude-sonnet-4.6",
1303
1307
  // $3/$15 - best for reasoning/instructions
1304
1308
  fallback: [
1305
- "claude-opus-4",
1306
- "claude-opus-4",
1309
+ "anthropic/claude-opus-4.6",
1310
+ "anthropic/claude-opus-4.6",
1307
1311
  "openai/o4-mini",
1308
1312
  // Newer and cheaper than o3 ($1.10 vs $2.00)
1309
1313
  "openai/o3",
@@ -1316,27 +1320,29 @@ var DEFAULT_ROUTING_CONFIG = {
1316
1320
  SIMPLE: {
1317
1321
  primary: "moonshot/kimi-k2.5",
1318
1322
  // Cheaper than Haiku ($0.5/$2.4 vs $1/$5), larger context
1319
- fallback: ["claude-haiku-4.5", "xai/grok-4-1-fast-non-reasoning", "openai/gpt-4o-mini"]
1323
+ fallback: ["anthropic/claude-haiku-4.5", "xai/grok-4-1-fast-non-reasoning", "openai/gpt-4o-mini"]
1320
1324
  },
1321
1325
  MEDIUM: {
1322
1326
  primary: "xai/grok-code-fast-1",
1323
1327
  // Code specialist for agentic coding
1324
- fallback: ["moonshot/kimi-k2.5", "claude-haiku-4.5", "claude-sonnet-4"]
1328
+ fallback: ["moonshot/kimi-k2.5", "anthropic/claude-haiku-4.5", "claude-sonnet-4"]
1325
1329
  },
1326
1330
  COMPLEX: {
1327
- primary: "claude-sonnet-4",
1331
+ primary: "anthropic/claude-sonnet-4.6",
1328
1332
  fallback: [
1329
- "claude-opus-4",
1333
+ "anthropic/claude-opus-4.6",
1330
1334
  // Latest Opus - best agentic
1331
1335
  "openai/gpt-5.2",
1336
+ "google/gemini-3.1-pro-preview",
1337
+ // Newest Gemini
1332
1338
  "google/gemini-3-pro-preview",
1333
1339
  "xai/grok-4-0709"
1334
1340
  ]
1335
1341
  },
1336
1342
  REASONING: {
1337
- primary: "claude-sonnet-4",
1343
+ primary: "anthropic/claude-sonnet-4.6",
1338
1344
  // Strong tool use + reasoning for agentic tasks
1339
- fallback: ["claude-opus-4", "xai/grok-4-1-fast-reasoning", "deepseek/deepseek-reasoner"]
1345
+ fallback: ["anthropic/claude-opus-4.6", "xai/grok-4-1-fast-reasoning", "deepseek/deepseek-reasoner"]
1340
1346
  }
1341
1347
  },
1342
1348
  overrides: {
@@ -1420,35 +1426,37 @@ function route(prompt, systemPrompt, maxOutputTokens, options) {
1420
1426
 
1421
1427
  // src/models.ts
1422
1428
  var MODEL_ALIASES = {
1423
- // Claude - short names (backend uses bare model names without anthropic/ prefix)
1424
- claude: "claude-sonnet-4",
1425
- sonnet: "claude-sonnet-4",
1426
- "sonnet-4.6": "claude-sonnet-4",
1427
- "sonnet-4-6": "claude-sonnet-4",
1428
- opus: "claude-opus-4",
1429
- "opus-4": "claude-opus-4",
1430
- haiku: "claude-haiku-4.5",
1429
+ // Claude - use newest versions (4.6)
1430
+ claude: "anthropic/claude-sonnet-4.6",
1431
+ sonnet: "anthropic/claude-sonnet-4.6",
1432
+ "sonnet-4": "anthropic/claude-sonnet-4.6",
1433
+ "sonnet-4.6": "anthropic/claude-sonnet-4.6",
1434
+ "sonnet-4-6": "anthropic/claude-sonnet-4.6",
1435
+ opus: "anthropic/claude-opus-4.6",
1436
+ "opus-4": "anthropic/claude-opus-4.6",
1437
+ "opus-4.6": "anthropic/claude-opus-4.6",
1438
+ "opus-4-6": "anthropic/claude-opus-4.6",
1439
+ haiku: "anthropic/claude-haiku-4.5",
1431
1440
  // Claude - provider/shortname patterns (common in agent frameworks)
1432
- "anthropic/sonnet": "claude-sonnet-4",
1433
- "anthropic/opus": "claude-opus-4",
1434
- "anthropic/haiku": "claude-haiku-4.5",
1435
- "anthropic/claude": "claude-sonnet-4",
1436
- // Backward compatibility - various formats all route to backend names
1437
- "anthropic/claude-sonnet-4": "claude-sonnet-4",
1438
- "anthropic/claude-sonnet-4-6": "claude-sonnet-4",
1439
- "anthropic/claude-sonnet-4.6": "claude-sonnet-4",
1440
- "anthropic/claude-opus-4": "claude-opus-4",
1441
- "anthropic/claude-opus-4-6": "claude-opus-4",
1442
- "anthropic/claude-opus-4.6": "claude-opus-4",
1443
- "anthropic/claude-haiku-4": "claude-haiku-4.5",
1444
- "anthropic/claude-haiku-4-5": "claude-haiku-4.5",
1445
- "anthropic/claude-haiku-4.5": "claude-haiku-4.5",
1441
+ "anthropic/sonnet": "anthropic/claude-sonnet-4.6",
1442
+ "anthropic/opus": "anthropic/claude-opus-4.6",
1443
+ "anthropic/haiku": "anthropic/claude-haiku-4.5",
1444
+ "anthropic/claude": "anthropic/claude-sonnet-4.6",
1445
+ // Backward compatibility - map all variants to 4.6
1446
+ "anthropic/claude-sonnet-4": "anthropic/claude-sonnet-4.6",
1447
+ "anthropic/claude-sonnet-4-6": "anthropic/claude-sonnet-4.6",
1448
+ "anthropic/claude-opus-4": "anthropic/claude-opus-4.6",
1449
+ "anthropic/claude-opus-4-6": "anthropic/claude-opus-4.6",
1450
+ "anthropic/claude-opus-4.5": "anthropic/claude-opus-4.6",
1451
+ "anthropic/claude-haiku-4": "anthropic/claude-haiku-4.5",
1452
+ "anthropic/claude-haiku-4-5": "anthropic/claude-haiku-4.5",
1446
1453
  // OpenAI
1447
1454
  gpt: "openai/gpt-4o",
1448
1455
  gpt4: "openai/gpt-4o",
1449
1456
  gpt5: "openai/gpt-5.2",
1450
1457
  codex: "openai/gpt-5.2-codex",
1451
1458
  mini: "openai/gpt-4o-mini",
1459
+ o1: "openai/o1",
1452
1460
  o3: "openai/o3",
1453
1461
  // DeepSeek
1454
1462
  deepseek: "deepseek/deepseek-chat",
@@ -1524,6 +1532,7 @@ var BLOCKRUN_MODELS = [
1524
1532
  {
1525
1533
  id: "openai/gpt-5.2",
1526
1534
  name: "GPT-5.2",
1535
+ version: "5.2",
1527
1536
  inputPrice: 1.75,
1528
1537
  outputPrice: 14,
1529
1538
  contextWindow: 4e5,
@@ -1535,6 +1544,7 @@ var BLOCKRUN_MODELS = [
1535
1544
  {
1536
1545
  id: "openai/gpt-5-mini",
1537
1546
  name: "GPT-5 Mini",
1547
+ version: "5.0",
1538
1548
  inputPrice: 0.25,
1539
1549
  outputPrice: 2,
1540
1550
  contextWindow: 2e5,
@@ -1543,6 +1553,7 @@ var BLOCKRUN_MODELS = [
1543
1553
  {
1544
1554
  id: "openai/gpt-5-nano",
1545
1555
  name: "GPT-5 Nano",
1556
+ version: "5.0",
1546
1557
  inputPrice: 0.05,
1547
1558
  outputPrice: 0.4,
1548
1559
  contextWindow: 128e3,
@@ -1551,6 +1562,7 @@ var BLOCKRUN_MODELS = [
1551
1562
  {
1552
1563
  id: "openai/gpt-5.2-pro",
1553
1564
  name: "GPT-5.2 Pro",
1565
+ version: "5.2",
1554
1566
  inputPrice: 21,
1555
1567
  outputPrice: 168,
1556
1568
  contextWindow: 4e5,
@@ -1561,8 +1573,9 @@ var BLOCKRUN_MODELS = [
1561
1573
  {
1562
1574
  id: "openai/gpt-5.2-codex",
1563
1575
  name: "GPT-5.2 Codex",
1564
- inputPrice: 2.5,
1565
- outputPrice: 12,
1576
+ version: "5.2",
1577
+ inputPrice: 1.75,
1578
+ outputPrice: 14,
1566
1579
  contextWindow: 128e3,
1567
1580
  maxOutput: 32e3,
1568
1581
  agentic: true
@@ -1571,6 +1584,7 @@ var BLOCKRUN_MODELS = [
1571
1584
  {
1572
1585
  id: "openai/gpt-4.1",
1573
1586
  name: "GPT-4.1",
1587
+ version: "4.1",
1574
1588
  inputPrice: 2,
1575
1589
  outputPrice: 8,
1576
1590
  contextWindow: 128e3,
@@ -1580,15 +1594,25 @@ var BLOCKRUN_MODELS = [
1580
1594
  {
1581
1595
  id: "openai/gpt-4.1-mini",
1582
1596
  name: "GPT-4.1 Mini",
1597
+ version: "4.1",
1583
1598
  inputPrice: 0.4,
1584
1599
  outputPrice: 1.6,
1585
1600
  contextWindow: 128e3,
1586
1601
  maxOutput: 16384
1587
1602
  },
1588
- // gpt-4.1-nano removed - replaced by gpt-5-nano
1603
+ {
1604
+ id: "openai/gpt-4.1-nano",
1605
+ name: "GPT-4.1 Nano",
1606
+ version: "4.1",
1607
+ inputPrice: 0.1,
1608
+ outputPrice: 0.4,
1609
+ contextWindow: 128e3,
1610
+ maxOutput: 16384
1611
+ },
1589
1612
  {
1590
1613
  id: "openai/gpt-4o",
1591
1614
  name: "GPT-4o",
1615
+ version: "4o",
1592
1616
  inputPrice: 2.5,
1593
1617
  outputPrice: 10,
1594
1618
  contextWindow: 128e3,
@@ -1599,15 +1623,37 @@ var BLOCKRUN_MODELS = [
1599
1623
  {
1600
1624
  id: "openai/gpt-4o-mini",
1601
1625
  name: "GPT-4o Mini",
1626
+ version: "4o-mini",
1602
1627
  inputPrice: 0.15,
1603
1628
  outputPrice: 0.6,
1604
1629
  contextWindow: 128e3,
1605
1630
  maxOutput: 16384
1606
1631
  },
1607
- // OpenAI O-series (Reasoning) - o1/o1-mini removed, replaced by o3/o4
1632
+ // OpenAI O-series (Reasoning)
1633
+ {
1634
+ id: "openai/o1",
1635
+ name: "o1",
1636
+ version: "1",
1637
+ inputPrice: 15,
1638
+ outputPrice: 60,
1639
+ contextWindow: 2e5,
1640
+ maxOutput: 1e5,
1641
+ reasoning: true
1642
+ },
1643
+ {
1644
+ id: "openai/o1-mini",
1645
+ name: "o1-mini",
1646
+ version: "1-mini",
1647
+ inputPrice: 1.1,
1648
+ outputPrice: 4.4,
1649
+ contextWindow: 128e3,
1650
+ maxOutput: 65536,
1651
+ reasoning: true
1652
+ },
1608
1653
  {
1609
1654
  id: "openai/o3",
1610
1655
  name: "o3",
1656
+ version: "3",
1611
1657
  inputPrice: 2,
1612
1658
  outputPrice: 8,
1613
1659
  contextWindow: 2e5,
@@ -1617,6 +1663,7 @@ var BLOCKRUN_MODELS = [
1617
1663
  {
1618
1664
  id: "openai/o3-mini",
1619
1665
  name: "o3-mini",
1666
+ version: "3-mini",
1620
1667
  inputPrice: 1.1,
1621
1668
  outputPrice: 4.4,
1622
1669
  contextWindow: 128e3,
@@ -1626,6 +1673,7 @@ var BLOCKRUN_MODELS = [
1626
1673
  {
1627
1674
  id: "openai/o4-mini",
1628
1675
  name: "o4-mini",
1676
+ version: "4-mini",
1629
1677
  inputPrice: 1.1,
1630
1678
  outputPrice: 4.4,
1631
1679
  contextWindow: 128e3,
@@ -1633,10 +1681,11 @@ var BLOCKRUN_MODELS = [
1633
1681
  reasoning: true
1634
1682
  },
1635
1683
  // Anthropic - all Claude models excel at agentic workflows
1636
- // Backend uses bare model names (claude-sonnet-4, not anthropic/claude-sonnet-4-6)
1684
+ // Use newest versions (4.6) with full provider prefix
1637
1685
  {
1638
- id: "claude-haiku-4.5",
1686
+ id: "anthropic/claude-haiku-4.5",
1639
1687
  name: "Claude Haiku 4.5",
1688
+ version: "4.5",
1640
1689
  inputPrice: 1,
1641
1690
  outputPrice: 5,
1642
1691
  contextWindow: 2e5,
@@ -1644,8 +1693,9 @@ var BLOCKRUN_MODELS = [
1644
1693
  agentic: true
1645
1694
  },
1646
1695
  {
1647
- id: "claude-sonnet-4",
1696
+ id: "anthropic/claude-sonnet-4.6",
1648
1697
  name: "Claude Sonnet 4.6",
1698
+ version: "4.6",
1649
1699
  inputPrice: 3,
1650
1700
  outputPrice: 15,
1651
1701
  contextWindow: 2e5,
@@ -1654,19 +1704,32 @@ var BLOCKRUN_MODELS = [
1654
1704
  agentic: true
1655
1705
  },
1656
1706
  {
1657
- id: "claude-opus-4",
1707
+ id: "anthropic/claude-opus-4.6",
1658
1708
  name: "Claude Opus 4.6",
1659
- inputPrice: 15,
1660
- outputPrice: 75,
1709
+ version: "4.6",
1710
+ inputPrice: 5,
1711
+ outputPrice: 25,
1661
1712
  contextWindow: 2e5,
1662
1713
  maxOutput: 32e3,
1663
1714
  reasoning: true,
1664
1715
  agentic: true
1665
1716
  },
1666
1717
  // Google
1718
+ {
1719
+ id: "google/gemini-3.1-pro-preview",
1720
+ name: "Gemini 3.1 Pro Preview",
1721
+ version: "3.1",
1722
+ inputPrice: 2,
1723
+ outputPrice: 12,
1724
+ contextWindow: 105e4,
1725
+ maxOutput: 65536,
1726
+ reasoning: true,
1727
+ vision: true
1728
+ },
1667
1729
  {
1668
1730
  id: "google/gemini-3-pro-preview",
1669
1731
  name: "Gemini 3 Pro Preview",
1732
+ version: "3.0",
1670
1733
  inputPrice: 2,
1671
1734
  outputPrice: 12,
1672
1735
  contextWindow: 105e4,
@@ -1674,9 +1737,20 @@ var BLOCKRUN_MODELS = [
1674
1737
  reasoning: true,
1675
1738
  vision: true
1676
1739
  },
1740
+ {
1741
+ id: "google/gemini-3-flash-preview",
1742
+ name: "Gemini 3 Flash Preview",
1743
+ version: "3.0",
1744
+ inputPrice: 0.5,
1745
+ outputPrice: 3,
1746
+ contextWindow: 1e6,
1747
+ maxOutput: 65536,
1748
+ vision: true
1749
+ },
1677
1750
  {
1678
1751
  id: "google/gemini-2.5-pro",
1679
1752
  name: "Gemini 2.5 Pro",
1753
+ version: "2.5",
1680
1754
  inputPrice: 1.25,
1681
1755
  outputPrice: 10,
1682
1756
  contextWindow: 105e4,
@@ -1687,8 +1761,18 @@ var BLOCKRUN_MODELS = [
1687
1761
  {
1688
1762
  id: "google/gemini-2.5-flash",
1689
1763
  name: "Gemini 2.5 Flash",
1690
- inputPrice: 0.15,
1691
- outputPrice: 0.6,
1764
+ version: "2.5",
1765
+ inputPrice: 0.3,
1766
+ outputPrice: 2.5,
1767
+ contextWindow: 1e6,
1768
+ maxOutput: 65536
1769
+ },
1770
+ {
1771
+ id: "google/gemini-2.5-flash-lite",
1772
+ name: "Gemini 2.5 Flash Lite",
1773
+ version: "2.5",
1774
+ inputPrice: 0.1,
1775
+ outputPrice: 0.4,
1692
1776
  contextWindow: 1e6,
1693
1777
  maxOutput: 65536
1694
1778
  },
@@ -1696,6 +1780,7 @@ var BLOCKRUN_MODELS = [
1696
1780
  {
1697
1781
  id: "deepseek/deepseek-chat",
1698
1782
  name: "DeepSeek V3.2 Chat",
1783
+ version: "3.2",
1699
1784
  inputPrice: 0.28,
1700
1785
  outputPrice: 0.42,
1701
1786
  contextWindow: 128e3,
@@ -1704,6 +1789,7 @@ var BLOCKRUN_MODELS = [
1704
1789
  {
1705
1790
  id: "deepseek/deepseek-reasoner",
1706
1791
  name: "DeepSeek V3.2 Reasoner",
1792
+ version: "3.2",
1707
1793
  inputPrice: 0.28,
1708
1794
  outputPrice: 0.42,
1709
1795
  contextWindow: 128e3,
@@ -1714,8 +1800,9 @@ var BLOCKRUN_MODELS = [
1714
1800
  {
1715
1801
  id: "moonshot/kimi-k2.5",
1716
1802
  name: "Kimi K2.5",
1717
- inputPrice: 0.5,
1718
- outputPrice: 2.4,
1803
+ version: "k2.5",
1804
+ inputPrice: 0.6,
1805
+ outputPrice: 3,
1719
1806
  contextWindow: 262144,
1720
1807
  maxOutput: 8192,
1721
1808
  reasoning: true,
@@ -1726,6 +1813,7 @@ var BLOCKRUN_MODELS = [
1726
1813
  {
1727
1814
  id: "xai/grok-3",
1728
1815
  name: "Grok 3",
1816
+ version: "3",
1729
1817
  inputPrice: 3,
1730
1818
  outputPrice: 15,
1731
1819
  contextWindow: 131072,
@@ -1736,6 +1824,7 @@ var BLOCKRUN_MODELS = [
1736
1824
  {
1737
1825
  id: "xai/grok-3-mini",
1738
1826
  name: "Grok 3 Mini",
1827
+ version: "3-mini",
1739
1828
  inputPrice: 0.3,
1740
1829
  outputPrice: 0.5,
1741
1830
  contextWindow: 131072,
@@ -1745,6 +1834,7 @@ var BLOCKRUN_MODELS = [
1745
1834
  {
1746
1835
  id: "xai/grok-4-fast-reasoning",
1747
1836
  name: "Grok 4 Fast Reasoning",
1837
+ version: "4",
1748
1838
  inputPrice: 0.2,
1749
1839
  outputPrice: 0.5,
1750
1840
  contextWindow: 131072,
@@ -1754,6 +1844,7 @@ var BLOCKRUN_MODELS = [
1754
1844
  {
1755
1845
  id: "xai/grok-4-fast-non-reasoning",
1756
1846
  name: "Grok 4 Fast",
1847
+ version: "4",
1757
1848
  inputPrice: 0.2,
1758
1849
  outputPrice: 0.5,
1759
1850
  contextWindow: 131072,
@@ -1762,6 +1853,7 @@ var BLOCKRUN_MODELS = [
1762
1853
  {
1763
1854
  id: "xai/grok-4-1-fast-reasoning",
1764
1855
  name: "Grok 4.1 Fast Reasoning",
1856
+ version: "4.1",
1765
1857
  inputPrice: 0.2,
1766
1858
  outputPrice: 0.5,
1767
1859
  contextWindow: 131072,
@@ -1771,6 +1863,7 @@ var BLOCKRUN_MODELS = [
1771
1863
  {
1772
1864
  id: "xai/grok-4-1-fast-non-reasoning",
1773
1865
  name: "Grok 4.1 Fast",
1866
+ version: "4.1",
1774
1867
  inputPrice: 0.2,
1775
1868
  outputPrice: 0.5,
1776
1869
  contextWindow: 131072,
@@ -1779,6 +1872,7 @@ var BLOCKRUN_MODELS = [
1779
1872
  {
1780
1873
  id: "xai/grok-code-fast-1",
1781
1874
  name: "Grok Code Fast",
1875
+ version: "1",
1782
1876
  inputPrice: 0.2,
1783
1877
  outputPrice: 1.5,
1784
1878
  contextWindow: 131072,
@@ -1789,17 +1883,28 @@ var BLOCKRUN_MODELS = [
1789
1883
  {
1790
1884
  id: "xai/grok-4-0709",
1791
1885
  name: "Grok 4 (0709)",
1886
+ version: "4-0709",
1792
1887
  inputPrice: 0.2,
1793
1888
  outputPrice: 1.5,
1794
1889
  contextWindow: 131072,
1795
1890
  maxOutput: 16384,
1796
1891
  reasoning: true
1797
1892
  },
1798
- // grok-2-vision removed - old, 0 transactions
1893
+ {
1894
+ id: "xai/grok-2-vision",
1895
+ name: "Grok 2 Vision",
1896
+ version: "2",
1897
+ inputPrice: 2,
1898
+ outputPrice: 10,
1899
+ contextWindow: 131072,
1900
+ maxOutput: 16384,
1901
+ vision: true
1902
+ },
1799
1903
  // MiniMax
1800
1904
  {
1801
1905
  id: "minimax/minimax-m2.5",
1802
1906
  name: "MiniMax M2.5",
1907
+ version: "m2.5",
1803
1908
  inputPrice: 0.3,
1804
1909
  outputPrice: 1.2,
1805
1910
  contextWindow: 204800,
@@ -1811,6 +1916,7 @@ var BLOCKRUN_MODELS = [
1811
1916
  {
1812
1917
  id: "nvidia/gpt-oss-120b",
1813
1918
  name: "NVIDIA GPT-OSS 120B",
1919
+ version: "120b",
1814
1920
  inputPrice: 0,
1815
1921
  outputPrice: 0,
1816
1922
  contextWindow: 128e3,
@@ -1819,6 +1925,7 @@ var BLOCKRUN_MODELS = [
1819
1925
  {
1820
1926
  id: "nvidia/kimi-k2.5",
1821
1927
  name: "NVIDIA Kimi K2.5",
1928
+ version: "k2.5",
1822
1929
  inputPrice: 0.55,
1823
1930
  outputPrice: 2.5,
1824
1931
  contextWindow: 262144,